Dissertations / Theses on the topic 'Artificial datasets'
To see the other types of publications on this topic, follow the link: Artificial datasets.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Artificial datasets.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Hilton, Erwin. "Visual datasets for artificial intelligence agents." Thesis, Massachusetts Institute of Technology, 2018. http://hdl.handle.net/1721.1/119553.
Abstract:
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from PDF version of thesis.
Includes bibliographical references (page 41).
In this thesis, I designed and implemented two visual dataset generation tool frameworks. With these tools, I introduce procedurally generated new data to test VQA agents and other visual Al models on. The first tool is Spatial IQ Generative Dataset (SIQGD). This tool generates images based on the Raven's Progressive Matrices spatial IQ examination metric. The second tool is a collection of 3D models along with a Blender3D extension that renders images of the models from multiple viewpoints along with their depth maps.
by Erwin Hilton.
M. Eng.
2
Siddique, Nahian A. "PATTERN RECOGNITION IN CLASS IMBALANCED DATASETS." VCU Scholars Compass, 2016. http://scholarscompass.vcu.edu/etd/4480.
Abstract:
Class imbalanced datasets constitute a significant portion of the machine learning problems of interest, where recognizing the ‘rare class’ is the primary objective for most applications. Traditional linear machine learning algorithms are often not effective in recognizing the rare class. In this research work, a specifically optimized feed-forward artificial neural network (ANN) is proposed and developed to train from moderate to highly imbalanced datasets.
The proposed methodology deals with the difficulty in classification task in multiple stages—by optimizing the training dataset, modifying kernel function to generate the gram matrix and optimizing the NN structure. First, the training dataset is extracted from the available sample set through an iterative process of selective under-sampling. Then, the proposed artificial NN comprises of a kernel function optimizer to specifically enhance class boundaries for imbalanced datasets by conformally transforming the kernel functions. Finally, a single hidden layer weighted neural network structure is proposed to train models from the imbalanced dataset. The proposed NN architecture is derived to effectively classify any binary dataset with even very high imbalance ratio with appropriate parameter tuning and sufficient number of processing elements.
Effectiveness of the proposed method is tested on accuracy based performance metrics, achieving close to and above 90%, with several imbalanced datasets of generic nature and compared with state of the art methods. The proposed model is also used for classification of a 25GB computed tomographic colonography database to test its applicability for big data. Also the effectiveness of under-sampling, kernel optimization for training of the NN model from the modified kernel gram matrix representing the imbalanced data distribution is analyzed experimentally. Computation time analysis shows the feasibility of the system for practical purposes. This report is concluded with discussion of prospect of the developed model and suggestion for further development works in this direction.
3
Lundberg, Oskar. "Decentralized machine learning on massive heterogeneous datasets : A thesis about vertical federated learning." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-444639.
Abstract:
The need for a method to create a collaborative machine learning model which can utilize data from different clients, each with privacy constraints, has recently emerged. This is due to privacy restrictions, such as General Data Protection Regulation, together with the fact that machine learning models in general needs large size data to perform well. Google introduced federated learning in 2016 with the aim to address this problem. Federated learning can further be divided into horizontal and vertical federated learning, depending on how the data is structured at the different clients. Vertical federated learning is applicable when many different features is obtained on distributed computation nodes, where they can not be shared in between. The aim of this thesis is to identify the current state of the art methods in vertical federated learning, implement the most interesting ones and compare the results in order to draw conclusions of the benefits and drawbacks of the different methods. From the results of the experiments, a method called FedBCD shows very promising results where it achieves massive improvements in the number of communication rounds needed for convergence, at the cost of more computations at the clients. A comparison between synchronous and asynchronous approaches shows slightly better results for the synchronous approach in scenarios with no delay. Delay refers to slower performance in one of the workers, either due to lower computational resources or due to communication issues. In scenarios where an artificial delay is implemented, the asynchronous approach shows superior results due to its ability to continue training in the case of delays in one or several of the clients.
4
Horečný, Peter. "Metody segmentace obrazu s malými trénovacími množinami." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2020. http://www.nusl.cz/ntk/nusl-412996.
Abstract:
The goal of this thesis was to propose an image segmentation method, which is capable of effective segmentation process with small datasets. Recently published ODE neural network was used for this method, because its features should provide better generalization in case of tasks with only small datasets available. The proposed ODE-UNet network was created by combining UNet architecture with ODE neural network, while using benefits of both networks. ODE-UNet reached following results on ISBI dataset: Rand: 0,950272 and Info: 0,978061. These results are better than the ones received from UNet model, which was also tested in this thesis, but it has been proven that state of the art can not be outperformed using ODE neural networks. However, the advantages of ODE neural network over tested UNet architecture and other methods were confirmed, and there is still a room for improvement by extending this method.
5
Woods, Brent J. "Computer-Aided Detection of Malignant Lesions in Dynamic Contrast Enhanced MRI Breast and Prostate Cancer Datasets." The Ohio State University, 2008. http://rave.ohiolink.edu/etdc/view?acc_num=osu1218155270.
6
Yasarer, Hakan. "Decision making in engineering prediction systems." Diss., Kansas State University, 2013. http://hdl.handle.net/2097/16231.
Abstract:
Doctor of PhilosophyDepartment of Civil Engineering
Yacoub M. Najjar
Access to databases after the digital revolutions has become easier because large databases are progressively available. Knowledge discovery in these databases via intelligent data analysis technology is a relatively young and interdisciplinary field. In engineering applications, there is a demand for turning low-level data-based knowledge into a high-level type knowledge via the use of various data analysis methods. The main reason for this demand is that collecting and analyzing databases can be expensive and time consuming. In cases where experimental or empirical data are already available, prediction models can be used to characterize the desired engineering phenomena and/or eliminate unnecessary future experiments and their associated costs. Phenomena characterization, based on available databases, has been utilized via Artificial Neural Networks (ANNs) for more than two decades. However, there is a need to introduce new paradigms to improve the reliability of the available ANN models and optimize their predictions through a hybrid decision system. In this study, a new set of ANN modeling approaches/paradigms along with a new method to tackle partially missing data (Query method) are introduced for this purpose. The potential use of these methods via a hybrid decision making system is examined by utilizing seven available databases which are obtained from civil engineering applications. Overall, the new proposed approaches have shown notable prediction accuracy improvements on the seven databases in terms of quantified statistical accuracy measures. The proposed new methods are capable in effectively characterizing the general behavior of a specific engineering/scientific phenomenon and can be collectively used to optimize predictions with a reasonable degree of accuracy. The utilization of the proposed hybrid decision making system (HDMS) via an Excel-based environment can easily be utilized by the end user, to any available data-rich database, without the need for any excessive type of training.
7
Gusarov, Nikita. "Performances des modèles économétriques et de Machine Learning pour l’étude économique des choix discrets de consommation." Electronic Thesis or Diss., Université Grenoble Alpes, 2024. http://www.theses.fr/2024GRALE001.
Abstract:
Cette thèse est une étude interdisciplinaire de la modélisation discrète des choix, abordant à la fois les techniques d'économétrie et d'apprentissage automatique (ML) appliquées à la modélisation des choix individuels de consommation. La problématique découle de points de contact insuffisants entre les utilisateurs (économistes et ingénieurs) et les analystes des données, qui poursuivent différents objectifs, bien qu'ils utilisent des techniques similaires. Pour combler cet écart interdisciplinaire, ce travail propose un framework unifié pour l'analyse des performances du modèle. Il facilite la comparaison des techniques d'analyse des données sous différentes hypothèses et transformations.Le framework conçu convient à une variété de modèles économétriques et ML. Il aborde la tâche de comparaison des performances du point de vue de la procédure de recherche, incorporant toutes les étapes affectant potentiellement les perceptions des performances. Pour démontrer les capacités du framework, nous proposons une série de 3 études appliquées. Dans ces études, la performance du modèle est explorée face aux changements de: (1) la taille et l'équilibre de l'échantillon, résultant de la collecte de données; (2) les changements de la structure des préférences au sein de la population, reflétant des hypothèses comportementales incorrectes; et (3) la sélection du modèle, directement liée à la perception des performancesThis thesis is a cross-disciplinary study of discrete choice modeling, addressing both econometrics and machine learning (ML) techniques applied to individual choice modeling. The problematic arises from insufficient points of contact among users (economists and engineers) and data scientists, who pursue different objectives, although using similar techniques. To bridge this interdisciplinary gap, the PhD work proposes a unified framework for model performance analysis. It facilitates the comparison of data analysis techniques under varying assumptions and transformations.The designed framework is suitable for a variety of econometrics and ML models. It addresses the performance comparison task from the research procedure perspective, incorporating all the steps potentially affecting the performance perceptions. To demonstrate the framework’s capabilities we propose a series of 3 applied studies. In those studies the model performance is explored face to the changes in (1) sample size and balance, resulting from data collection; (2) changes in preferences structure within population, reflecting incorrect behavioral assumptions; and (3) model selection, directly intertwined with the performance perception
8
Matsumoto, Élia Yathie. "A methodology for improving computed individual regressions predictions." Universidade de São Paulo, 2015. http://www.teses.usp.br/teses/disponiveis/3/3142/tde-12052016-140407/.
Abstract:
This research proposes a methodology to improve computed individual prediction values provided by an existing regression model without having to change either its parameters or its architecture. In other words, we are interested in achieving more accurate results by adjusting the calculated regression prediction values, without modifying or rebuilding the original regression model. Our proposition is to adjust the regression prediction values using individual reliability estimates that indicate if a single regression prediction is likely to produce an error considered critical by the user of the regression. The proposed method was tested in three sets of experiments using three different types of data. The first set of experiments worked with synthetically produced data, the second with cross sectional data from the public data source UCI Machine Learning Repository and the third with time series data from ISO-NE (Independent System Operator in New England). The experiments with synthetic data were performed to verify how the method behaves in controlled situations. In this case, the outcomes of the experiments produced superior results with respect to predictions improvement for artificially produced cleaner datasets with progressive worsening with the addition of increased random elements. The experiments with real data extracted from UCI and ISO-NE were done to investigate the applicability of the methodology in the real world. The proposed method was able to improve regression prediction values by about 95% of the experiments with real data.Esta pesquisa propõe uma metodologia para melhorar previsões calculadas por um modelo de regressão, sem a necessidade de modificar seus parâmetros ou sua arquitetura. Em outras palavras, o objetivo é obter melhores resultados por meio de ajustes nos valores computados pela regressão, sem alterar ou reconstruir o modelo de previsão original. A proposta é ajustar os valores previstos pela regressão por meio do uso de estimadores de confiabilidade individuais capazes de indicar se um determinado valor estimado é propenso a produzir um erro considerado crítico pelo usuário da regressão. O método proposto foi testado em três conjuntos de experimentos utilizando três tipos de dados diferentes. O primeiro conjunto de experimentos trabalhou com dados produzidos artificialmente, o segundo, com dados transversais extraídos no repositório público de dados UCI Machine Learning Repository, e o terceiro, com dados do tipo séries de tempos extraídos do ISO-NE (Independent System Operator in New England). Os experimentos com dados artificiais foram executados para verificar o comportamento do método em situações controladas. Nesse caso, os experimentos alcançaram melhores resultados para dados limpos artificialmente produzidos e evidenciaram progressiva piora com a adição de elementos aleatórios. Os experimentos com dados reais extraído das bases de dados UCI e ISO-NE foram realizados para investigar a aplicabilidade da metodologia no mundo real. O método proposto foi capaz de melhorar os valores previstos por regressões em cerca de 95% dos experimentos realizados com dados reais.
9
Gualandi, Giacomo. "Analisi di dataset in campo finanziario mediante reti neurali LSTM." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2019. http://amslaurea.unibo.it/19623/.
Abstract:
Con il presente elaborato si è esplorato il campo della data analytics.
È stato analizzato un dataset relativo all' andamento storico del titolo di borsa di una società, i cui dati sono stati manipolati in modo tale da renderli compatibili per un loro utilizzo in una applicazione di Machine Learning. Si sono approfondite le reti neurali artificiali LSTM e con esse si è creato un modello che permettesse di effettuare delle predizioni sui valori futuri del titolo. Infine sono state valutate le differenze tra i valori predetti e quelli reali assunti dal titolo di borsa.
10
Mattiussi, Vlad. "Una Rassegna di Dataset e Applicazioni Innovative di Intelligenza Artificiale per Affrontare la Pandemia da COVID19." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/21844/.
Abstract:
Il machine learning e la computer vision hanno avuto rilevanti sviluppi negli ultimi anni, compiendo progressi in molti settori. L’IA ha contribuito ad affrontare la pandemia di coronavirus (COVID-19). La scienza e la tecnologia hanno contribuito in modo significativo all’attuazione di queste politiche in questo caotico periodo senza precedenti. Ad esempio, i robot vengono utilizzati negli ospedali per fornire cibo e medicine ai pazienti con coronavirus o i droni vengono utilizzati per disinfettare strade e spazi pubblici I ricercatori di informatica, d’altra parte, sono riusciti a rilevare precocemente i pazienti infettivi utilizzando tecniche in grado di elaborare e comprendere i dati di imaging medico come immagini a raggi X e scansioni di tomografia computerizzata (CT). Tutte queste tecniche computazionali fanno parte dell’intelligenza artificiale, che è stata applicata con successo in vari campi.
11
Nett, Ryan. "Dataset and Evaluation of Self-Supervised Learning for Panoramic Depth Estimation." DigitalCommons@CalPoly, 2020. https://digitalcommons.calpoly.edu/theses/2234.
Abstract:
Depth detection is a very common computer vision problem. It shows up primarily in robotics, automation, or 3D visualization domains, as it is essential for converting images to point clouds. One of the poster child applications is self driving cars. Currently, the best methods for depth detection are either very expensive, like LIDAR, or require precise calibration, like stereo cameras. These costs have given rise to attempts to detect depth from a monocular camera (a single camera). While this is possible, it is harder than LIDAR or stereo methods since depth can't be measured from monocular images, it has to be inferred. A good example is covering one eye: you still have some idea how far away things are, but it's not exact. Neural networks are a natural fit for this. Here, we build on previous neural network methods by applying a recent state of the art model to panoramic images in addition to pinhole ones and performing a comparative evaluation. First, we create a simulated depth detection dataset that lends itself to panoramic comparisons and contains pre-made cylindrical and spherical panoramas. We then modify monodepth2 to support cylindrical and cubemap panoramas, incorporating current best practices for depth detection on those panorama types, and evaluate its performance for each type of image using our dataset. We also consider the resources used in training and other qualitative factors.
12
Elatfi, Hamza. "Sviluppo di un sistema di crowdsourcing per la validazione e l'arricchimento di dataset." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2019. http://amslaurea.unibo.it/18452/.
Abstract:
La tesi tratta nozioni base degli argomenti di intelligenza artificiale, machine learning e crowdsourcing. Questi nozioni saranno utili a comprendere l'argomento centrale della tesi, ovvero, la creazione di un software che permetta: la raccolta di annotazioni effettuate dagli utenti e la creazione da parte di quest'ultimi di nuove entità.
13
Bartocci, John Timothy. "Generating a synthetic dataset for kidney transplantation using generative adversarial networks and categorical logit encoding." Bowling Green State University / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1617104572023027.
14
Lohniský, Michal. "Všesměrová detekce objektů." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2014. http://www.nusl.cz/ntk/nusl-236093.
Abstract:
This thesis focuses on modification of feature extraction and multiview object detection learning process. We add new channels to detectors based on the "Aggregate channel features" framework. These new channels are created by filtering the picture by kernels from autoencoders followed by nonlinear function processing. Experiments show that these channels are effective in detection but they are also more computationally expensive. The thesis therefore discusses possibilities for improvements. Finally the thesis evaluates an artificial car dataset and discusses its small benefit on several detectors.
15
Bianchi, Eric Loran. "COCO-Bridge: Common Objects in Context Dataset and Benchmark for Structural Detail Detection of Bridges." Thesis, Virginia Tech, 2019. http://hdl.handle.net/10919/87588.
Abstract:
Common Objects in Context for bridge inspection (COCO-Bridge) was introduced for use by unmanned aircraft systems (UAS) to assist in GPS denied environments, flight-planning, and detail identification and contextualization, but has far-reaching applications such as augmented reality (AR) and other artificial intelligence (AI) platforms. COCO-Bridge is an annotated dataset which can be trained using a convolutional neural network (CNN) to identify specific structural details. Many annotated datasets have been developed to detect regions of interest in images for a wide variety of applications and industries. While some annotated datasets of structural defects (primarily cracks) have been developed, most efforts are individualized and focus on a small niche of the industry. This effort initiated a benchmark dataset with a focus on structural details. This research investigated the required parameters for detail identification and evaluated performance enhancements on the annotation process. The image dataset consisted of four structural details which are commonly reviewed and rated during bridge inspections: bearings, cover plate terminations, gusset plate connections, and out of plane stiffeners. This initial version of COCO-Bridge includes a total of 774 images; 10% for evaluation and 90% for training. Several models were used with the dataset to evaluate model overfitting and performance enhancements from augmentation and number of iteration steps. Methods to economize the predictive capabilities of the model without the addition of unique data were investigated to reduce the required number of training images. Results from model tests indicated the following: additional images, mirrored along the vertical-axis, provided precision and accuracy enhancements; increasing computational step iterations improved predictive precision and accuracy, and the optimal confidence threshold for operation was 25%. Annotation recommendations and improvements were also discovered and documented as a result of the research.MS
Common Objects in Context for bridge inspection (COCO-Bridge) was introduced to improve a drone-conducted bridge inspection process. Drones are a great tool for bridge inspectors because they bring flexibility and access to the inspection. However, drones have a notoriously difficult time operating near bridges, because the signal can be lost between the operator and the drone. COCO-Bridge is an imagebased dataset that uses Artificial Intelligence (AI) as a solution to this particular problem, but has applications in other facets of the inspection as well. This effort initiated a dataset with a focus on identifying specific parts of a bridge or structural bridge elements. This would allow a drone to fly without explicit direction if the signal was lost, and also has the potential to extend its flight time. Extending flight time and operating autonomously are great advantagesfor drone operators and bridge inspectors. The output from COCO-Bridge would also help the inspectors identify areas that are prone to defects by highlighting regions that require inspection. The image dataset consisted of 774 images to detect four structural bridge elements which are commonly reviewed and rated during bridge inspections. The goal is to continue to increase the number of images and encompass more structural bridge elements in the dataset so that it may be used for all types of bridges. Methods to reduce the required number of images were investigated, because gathering images of structural bridge elements is challenging,. The results from model tests helped build a roadmap for the expansion and best-practices for developing a dataset of this type.
16
Poggi, Cavalletti Stefano. "Utilizzo di tecniche di Machine Learning per l'analisi di dataset in ambito sanitario." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/21743/.
Abstract:
L’intelligenza artificiale è una disciplina molto vasta che presenta numerosi campi di applicazione. In questo lavoro si vuole mostrare come sia possibile utilizzare alcune tecniche di Machine e Deep Learning per l’analisi di dataset nell'ambito sanitario. In particolare, dopo una prima introduzione all’argomento e ai principali algoritmi di apprendimento, viene analizzato un framework caratterizzato da un approccio orientato all’ingegneria del software che utilizza tecniche di machine learning per migliorare l’efficienza dei sistemi sanitari. Vengono poi descritte le varie fasi di un esperimento svolto che consiste nell’analisi di un dataset e la successiva creazione di un modello di classificazione per la predizione di malattie cardiache nei pazienti, utilizzando le reti neurali artificiali.
17
Hou, Chuanchuan. "Vibration-based damage identification with enhanced frequency dataset and a cracked beam element model." Thesis, University of Edinburgh, 2016. http://hdl.handle.net/1842/20434.
Abstract:
Damage identification is an important topic in structural assessment and structural health monitoring (SHM). Vibration-based identification techniques use modal data to identify the existence, location and severity of possible damages in structures, often via a numerical model updating procedure. Among other factors influencing the practicality and reliability of a damage identification approach, two are of primary interest to this study. The first one concerns the amount and quality of modal data that can be used as ‘response’ data for the model updating. It is generally recognised that natural frequencies can be measured with relatively high accuracy; however, their number is limited. Mode shapes, on the other hand, are susceptible to larger measurement errors. Seeking additional modal frequency data is therefore of significant value. The second one concerns the errors at the numerical (finite element) model level, particularly in the representation of the effect of damage on the dynamic properties of the structure. An inadequate damage model can lead to inaccurate and even false damage identification. The first part of the thesis is devoted to enhancing the modal dataset by extracting the so called ‘artificial boundary condition’ (ABC) frequencies in a real measurement environment. The ABC frequencies correspond to the natural frequencies of the structure with a perturbed boundary condition, but can be generated without the need of actually altering the physical support condition. A comprehensive experimental study on the extraction of such frequencies has been conducted. The test specimens included steel beams of relatively flexible nature, as well as thick and stiffer beams made from metal material and reinforced concrete, to cover the typical variation of the dynamic characteristics of real-life structures in a laboratory condition. The extracted ABC frequencies are subsequently applied in the damage identification in beams. Results demonstrate that it is possible to extract the first few ABC frequencies from the modal testing in different beam settings for a variety of ABC incorporating one or two virtual pin supports. The inclusion of ABC frequencies enables the identification of structural damages satisfactorily without the necessity to involve the mode shape information. The second part of the thesis is devoted to developing a robust model updating and damage identification approach for beam cracks, with a special focus on thick beams which present a more challenging problem in terms of the effect of a crack than slender beams. The priority task has been to establish a crack model which comprehensively describes the effect of a crack to reduce the modelling errors. A cracked Timoshenko beam element model is introduced for explicit beam crack identification. The cracked beam element model is formulated by incorporating an additional flexibility due to a crack using the fracture mechanics principles. Complex effects in cracked thick beams, including shear deformation and coupling between transverse and longitudinal vibrations, are represented in the model. The accuracy of the cracked beam element model for predicting modal data of cracked thick beams is first verified against numerically simulated examples. The consistency of predictions across different modes is examined in comparison with the conventional stiffness reduction approach. Upon satisfactory verification, a tailored model updating procedure incorporating an adaptive discretisation approach is developed for the implementation of the cracked beam element model for crack identification. The updating procedure is robust in that it has no restriction on the location, severity and number of cracks to be identified. Example updating results demonstrate that satisfactory identification can be achieved for practically any configurations of cracks in a beam. Experimental study with five solid beam specimens is then carried out to further verify the developed cracked beam element model. Both forward verification and crack damage identification with the tested beams show similar level of accuracy to that with the numerically simulated examples. The cracked beam element model can be extended to crack identification of beams with complex cross sections. To do so the additional flexibility matrix for a specific cross-section type needs to be re-formulated. In the present study this is done for box sections. The stress intensity factors (SIF) for a box section as required for the establishment of the additional flexibility matrix are formulated with an empirical approach combining FE simulation, parametric analysis and regression analysis. The extended cracked beam element model is verified against both FE simulated and experimentally measured modal data. The model is subsequently incorporated in the crack identification for box beams. The successful extension of the cracked beam element model to the box beams paves the way for similar extension to the crack identification of other types of sections in real-life engineering applications.
18
Furundzic, Bojan, and Fabian Mathisson. "Dataset Evaluation Method for Vehicle Detection Using TensorFlow Object Detection API." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-43345.
Abstract:
Recent developments in the field of object detection have highlighted a significant variation in quality between visual datasets. As a result, there is a need for a standardized approach of validating visual dataset features and their performance contribution. With a focus on vehicle detection, this thesis aims to develop an evaluation method utilized for comparing visual datasets. This method was utilized to determine the dataset that contributed to the detection model with the greatest ability to detect vehicles. The visual datasets compared in this research were BDD100K, KITTI and Udacity, each one being trained on individual models. Applying the developed evaluation method, a strong indication of BDD100K's performance superiority was determined. Further analysis and feature extraction of dataset size, label distribution and average labels per image was conducted. In addition, real-world experimental conduction was performed in order to validate the developed evaluation method. It could be determined that all features and experimental results pointed to BDD100K's superiority over the other datasets, validating the developed evaluation method. Furthermore, the TensorFlow Object Detection API's ability to improve performance gain from a visual dataset was studied. Through the use of augmentations, it was concluded that the TensorFlow Object Detection API serves as a great tool to increase performance gain for visual datasets.Inom fältet av objektdetektering har ny utveckling demonstrerat stor kvalitetsvariation mellan visuella dataset. Till följd av detta finns det ett behov av standardiserade valideringsmetoder för att jämföra visuella dataset och deras prestationsförmåga. Detta examensarbete har, med ett fokus på fordonsigenkänning, som syfte att utveckla en pålitlig valideringsmetod som kan användas för att jämföra visuella dataset. Denna valideringsmetod användes därefter för att fastställa det dataset som bidrog till systemet med bäst förmåga att detektera fordon. De dataset som användes i denna studien var BDD100K, KITTI och Udacity, som tränades på individuella igenkänningsmodeller. Genom att applicera denna valideringsmetod, fastställdes det att BDD100K var det dataset som bidrog till systemet med bäst presterande igenkänningsförmåga. En analys av dataset storlek, etikettdistribution och genomsnittliga antalet etiketter per bild var även genomförd. Tillsammans med ett experiment som genomfördes för att testa modellerna i verkliga sammanhang, kunde det avgöras att valideringsmetoden stämde överens med de fastställda resultaten. Slutligen studerades TensorFlow Object Detection APIs förmåga att förbättra prestandan som erhålls av ett visuellt dataset. Genom användning av ett modifierat dataset, kunde det fastställas att TensorFlow Object Detection API är ett lämpligt modifieringsverktyg som kan användas för att öka prestandan av ett visuellt dataset.
19
Thumé, Gabriela Salvador. "Geração de imagens artificiais e quantização aplicadas a problemas de classificação." Universidade de São Paulo, 2016. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-16122016-150334/.
Abstract:
Cada imagem pode ser representada como uma combinação de diversas características, como por exemplo o histograma de intensidades de cor ou propriedades de textura da imagem. Essas características compõem um vetor multidimensional que representa a imagem. É comum esse vetor ser dado como entrada para um método de classificação de padrões que, após aprender por meio de diversos exemplos, pode gerar um modelo de decisão. Estudos sugerem evidências de que a preparação das imagens-- por meio da especificação cuidadosa da aquisição, pré-processamento e segmentação-- pode impactar significativamente a classificação. Além da falta de tratamento das imagens antes da extração de características, o desbalanceamento de classes também se apresenta como um obstáculo para que a classificação seja satisfatória. Imagens possuem características que podem ser exploradas para melhorar a descrição dos objetos de interesse e, portanto, sua classificação. Entre as possibilidades de melhorias estão: a redução do número de intensidades das imagens antes da extração de características ao invés de métodos de quantização no vetor já extraído; e a geração de imagens a partir das originais, de forma a promover o balanceamento de bases de dados cujo número de exemplos de cada classe é desbalanceado. Portanto, a proposta desta dissertação é melhorar a classificação de imagens utilizando métodos de processamento de imagens antes da extração de características. Especificamente, busca analisar a influência do balanceamento de bases de dados e da quantização na classificação. Este estudo analisa ainda a visualização do espaço de características após os métodos de geração artificial de imagens e de interpolação das características extraídas das imagens originais (SMOTE), comparando como espaço original. A ênfase dessa visualização se dá na observação da importância do rebalanceamento das classes. Os resultados obtidos indicam que a quantização simplifica as imagens antes da extração de características e posterior redução de dimensionalidade, produzindo vetores mais compactos; e que o rebalanceamento de classes de imagens através da geração de imagens artificiais pode melhorar a classificação da base de imagens, em relação à classificação original e ao uso de métodos no espaço de características já extraídas.Each image can be represented by a combination of several features like color frequency and texture properties. Those features compose a multidimensional vector, which represents the original image. Commonly this vector is given as an input to a classification method that can learn from examplesand build a decision model. The literature suggests that image preparation steps like acute acquisition, preprocessing and segmentation can positively impact such classification. Besides that, class unbalancing is also a barrier to achieve good classification accuracy. Some features and methods can be explored to improveobjects\' description, thus their classification. Possible suggestions include: reducing colors number before feature extraction instead of applying quantization methods to raw vectors already extracted; and generating synthetic images from original ones, to balance the number of samples in an uneven data set. We propose to improve image classification using image processing methods before feature extraction. Specifically we want to analyze the influence of both balancing and quantization methods while applied to datasets in a classification routine. This research also analyses the visualization of feature space after the artificial image generation and feature interpolation (SMOTE), against to original space. Such visualization is used because it allows us to know how important is the rebalacing method. The results show that quantization simplifies imagesby producing compacted vectors before feature extraction and dimensionality reduction; and that using artificial generation to rebalance image datasets can improve classification, when compared to the original one and to applying methods on the already extracted feature vectors.
20
Malazizi, Ladan. "Development of Artificial Intelligence-based In-Silico Toxicity Models. Data Quality Analysis and Model Performance Enhancement through Data Generation." Thesis, University of Bradford, 2008. http://hdl.handle.net/10454/4262.
Abstract:
Toxic compounds, such as pesticides, are routinely tested against a range of aquatic, avian and mammalian species as part of the registration process. The need for reducing dependence on animal testing has led to an increasing interest in alternative methods such as in silico modelling. The QSAR (Quantitative Structure Activity Relationship)-based models are already in use for predicting physicochemical properties, environmental fate, eco-toxicological effects, and specific biological endpoints for a wide range of chemicals. Data plays an important role in modelling QSARs and also in result analysis for toxicity testing processes. This research addresses number of issues in predictive toxicology. One issue is the problem of data quality. Although large amount of toxicity data is available from online sources, this data may contain some unreliable samples and may be defined as of low quality. Its presentation also might not be consistent throughout different sources and that makes the access, interpretation and comparison of the information difficult. To address this issue we started with detailed investigation and experimental work on DEMETRA data. The DEMETRA datasets have been produced by the EC-funded project DEMETRA. Based on the investigation, experiments and the results obtained, the author identified a number of data quality criteria in order to provide a solution for data evaluation in toxicology domain. An algorithm has also been proposed to assess data quality before modelling. Another issue considered in the thesis was the missing values in datasets for toxicology domain. Least Square Method for a paired dataset and Serial Correlation for single version dataset provided the solution for the problem in two different situations. A procedural algorithm using these two methods has been proposed in order to overcome the problem of missing values. Another issue we paid attention to in this thesis was modelling of multi-class data sets in which the severe imbalance class samples distribution exists. The imbalanced data affect the performance of classifiers during the classification process. We have shown that as long as we understand how class members are constructed in dimensional space in each cluster we can reform the distribution and provide more knowledge domain for the classifier.
21
La, Mura Francesco. "Tecniche di Preparazione di Dataset da Immagini Satellitari di Siti Archeologici per Elaborazioni con Deep Learning." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/20428/.
Abstract:
Questa tesi riporta il contributo dato ad un progetto di applicazione di deep learning al remote sensing nell’ambito dell’archeologia. Il progetto ha come obiettivo quello di sviluppare una rete neurale, che sia in grado di analizzare una zona di interesse degli archeologi data in input, per dare come output una serie di possibili siti archeologici. L’area da cui provengono le immagini che verranno usate nella creazione del dataset è il governatorato di al-Q ̄adisiyya, che si trova in Iraq. Il progetto si inserisce come una delle tante iniziative portate avanti dall’Università di Bologna nel medioriente, infatti in ambito archeologico l’Alma Mater porta avanti due progetti molto importanti: Eduu dedicato alla divulgazione del patrimonio archeologico e culturale iracheno, e Qadis, pensato per la ricognizione mirata di siti archeologici con tecnologie avanzate. Per gli archeologi la ricerca dei siti di scavo tramite l’uso di tecniche di remote sensing, è un’attività dispendiosa a livello temporale. Da qui nasce il desiderio di tentare l’automazione di questo processo, o per lo meno di fornire un supporto automatizzato all’umano,così da poter meglio spendere tempo ed energie su altre attività relative alla professione di archeologo.
In questo contesto, lo scopo del lavoro di tesi è quello di collaborare al progetto con la creazione di un dataset, che sia adatto per l’allenamento di un modello direte neurale, che sarà poi usato per automatizzare il processo di individuazione, si presume che automatizzare questo processo possa aiutare gli archeologi a migliorare ancora di più la loro produttività con un risparmio di tempo e risorse. Un altro obiettivo è quello di abbattere la barriera di significato tra archeologi e data scientist, permettendo ad entrambi di collaborare nella creazione di un modello di rete neurale, che sfrutti le conoscenze umane per ottenere risultati migliori rispetto ad un modello allenato senza alcun interazione con gli esperti del settore.
22
Sievert, Rolf. "Instance Segmentation of Multiclass Litter and Imbalanced Dataset Handling : A Deep Learning Model Comparison." Thesis, Linköpings universitet, Datorseende, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-175173.
Abstract:
Instance segmentation has a great potential for improving the current state of littering by autonomously detecting and segmenting different categories of litter. With this information, litter could, for example, be geotagged to aid litter pickers or to give precise locational information to unmanned vehicles for autonomous litter collection. Land-based litter instance segmentation is a relatively unexplored field, and this study aims to give a comparison of the instance segmentation models Mask R-CNN and DetectoRS using the multiclass litter dataset called Trash Annotations in Context (TACO) in conjunction with the Common Objects in Context precision and recall scores. TACO is an imbalanced dataset, and therefore imbalanced data-handling is addressed, exercising a second-order relation iterative stratified split, and additionally oversampling when training Mask R-CNN. Mask R-CNN without oversampling resulted in a segmentation of 0.127 mAP, and with oversampling 0.163 mAP. DetectoRS achieved 0.167 segmentation mAP, and improves the segmentation mAP of small objects most noticeably, with a factor of at least 2, which is important within the litter domain since small objects such as cigarettes are overrepresented. In contrast, oversampling with Mask R-CNN does not seem to improve the general precision of small and medium objects, but only improves the detection of large objects. It is concluded that DetectoRS improves results compared to Mask R-CNN, as well does oversampling. However, using a dataset that cannot have an all-class representation for train, validation, and test splits, together with an iterative stratification that does not guarantee all-class representations, makes it hard for future works to do exact comparisons to this study. Results are therefore approximate considering using all categories since 12 categories are missing from the test set, where 4 of those were impossible to split into train, validation, and test set. Further image collection and annotation to mitigate the imbalance would most noticeably improve results since results depend on class-averaged values. Doing oversampling with DetectoRS would also help improve results. There is also the option to combine the two datasets TACO and MJU-Waste to enforce training of more categories.
23
Nilsson, Alexander, and Martin Thönners. "A Framework for Generative Product Design Powered by Deep Learning and Artificial Intelligence : Applied on Everyday Products." Thesis, Linköpings universitet, Maskinkonstruktion, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-149454.
Abstract:
In this master’s thesis we explore the idea of using artificial intelligence in the product design process and seek to develop a conceptual framework for how it can be incorporated to make user customized products more accessible and affordable for everyone. We show how generative deep learning models such as Variational Auto Encoders and Generative Adversarial Networks can be implemented to generate design variations of windows and clarify the general implementation process along with insights from recent research in the field. The proposed framework consists of three parts: (1) A morphological matrix connecting several identified possibilities of implementation to specific parts of the product design process. (2) A general step-by-step process on how to incorporate generative deep learning. (3) A description of common challenges, strategies andsolutions related to the implementation process. Together with the framework we also provide a system for automatic gathering and cleaning of image data as well as a dataset containing 4564 images of windows in a front view perspective.
24
Barbazza, Sigfrido. "Deep-learning applicato all'identificazione automatica di frutta in immagini." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2016. http://amslaurea.unibo.it/11526/.
Abstract:
Analisi delle fasi per la realizzazione di uno strumento di supporto gli agricoltori, dalla creazione di un dataset, all'addestramento e test di una rete neurale artificiale, con obiettivo la localizzazione del prodotto agricolo all'interno delle immagini.
25
Johansson, David. "Price Prediction of Vinyl Records Using Machine Learning Algorithms." Thesis, Linnéuniversitetet, Institutionen för datavetenskap och medieteknik (DM), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-96464.
Abstract:
Machine learning algorithms have been used for price prediction within several application areas. Examples include real estate, the stock market, tourist accommodation, electricity, art, cryptocurrencies, and fine wine. Common approaches in studies are to evaluate the accuracy of predictions and compare different algorithms, such as Linear Regression or Neural Networks. There is a thriving global second-hand market for vinyl records, but the research of price prediction within the area is very limited. The purpose of this project was to expand on existing knowledge within price prediction in general to evaluate some aspects of price prediction of vinyl records. That included investigating the possible level of accuracy and comparing the efficiency of algorithms. A dataset of 37000 samples of vinyl records was created with data from the Discogs website, and multiple machine learning algorithms were utilized in a controlled experiment. Among the conclusions drawn from the results was that the Random Forest algorithm generally generated the strongest results, that results can vary substantially between different artists or genres, and that a large part of the predictions had a good accuracy level, but that a relatively small amount of large errors had a considerable effect on the general results.
26
Udaya, Kumar Magesh Kumar. "Classification of Parkinson’s Disease using MultiPass Lvq,Logistic Model Tree,K-Star for Audio Data set : Classification of Parkinson Disease using Audio Dataset." Thesis, Högskolan Dalarna, Datateknik, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:du-5596.
Abstract:
Parkinson's disease (PD) is a degenerative illness whose cardinal symptoms include rigidity, tremor, and slowness of movement. In addition to its widely recognized effects PD can have a profound effect on speech and voice.The speech symptoms most commonly demonstrated by patients with PD are reduced vocal loudness, monopitch, disruptions of voice quality, and abnormally fast rate of speech. This cluster of speech symptoms is often termed Hypokinetic Dysarthria.The disease can be difficult to diagnose accurately, especially in its early stages, due to this reason, automatic techniques based on Artificial Intelligence should increase the diagnosing accuracy and to help the doctors make better decisions. The aim of the thesis work is to predict the PD based on the audio files collected from various patients.Audio files are preprocessed in order to attain the features.The preprocessed data contains 23 attributes and 195 instances. On an average there are six voice recordings per person, By using data compression technique such as Discrete Cosine Transform (DCT) number of instances can be minimized, after data compression, attribute selection is done using several WEKA build in methods such as ChiSquared, GainRatio, Infogain after identifying the important attributes, we evaluate attributes one by one by using stepwise regression.Based on the selected attributes we process in WEKA by using cost sensitive classifier with various algorithms like MultiPass LVQ, Logistic Model Tree(LMT), K-Star.The classified results shows on an average 80%.By using this features 95% approximate classification of PD is acheived.This shows that using the audio dataset, PD could be predicted with a higher level of accuracy.
27
Galassi, Andrea. "Symbolic versus sub-symbolic approaches: a case study on training Deep Networks to play Nine Men’s Morris game." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2017. http://amslaurea.unibo.it/12859/.
Abstract:
Le reti neurali artificiali, grazie alle nuove tecniche di Deep Learning, hanno completamente rivoluzionato il panorama tecnologico degli ultimi anni, dimostrandosi efficaci in svariati compiti di Intelligenza Artificiale e ambiti affini. Sarebbe quindi interessante analizzare in che modo e in quale misura le deep network possano sostituire le IA simboliche. Dopo gli impressionanti risultati ottenuti nel gioco del Go, come caso di studio è stato scelto il gioco del Mulino, un gioco da tavolo largamente diffuso e ampiamente studiato. È stato quindi creato il sistema completamente sub-simbolico Neural Nine Men’s Morris, che sfrutta tre reti neurali per scegliere la mossa migliore. Le reti sono state addestrate su un dataset di più di 1.500.000 coppie (stato del gioco, mossa migliore), creato in base alle scelte di una IA simbolica. Il sistema ha dimostrato di aver imparato le regole del gioco proponendo una mossa valida in più del 99% dei casi di test. Inoltre ha raggiunto un’accuratezza del 39% rispetto al dataset e ha sviluppato una propria strategia di gioco diversa da quella della IA addestratrice, dimostrandosi un giocatore peggiore o migliore a seconda dell’avversario. I risultati ottenuti in questo caso di studio mostrano che, in questo contesto, la chiave del successo nella progettazione di sistemi AI allo stato dell’arte sembra essere un buon bilanciamento tra tecniche simboliche e sub-simboliche, dando più rilevanza a queste ultime, con lo scopo di raggiungere la perfetta integrazione di queste tecnologie.
28
Kaster, Joshua M. "Training Convolutional Neural Network Classifiers Using Simultaneous Scaled Supercomputing." University of Dayton / OhioLINK, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1588973772607826.
29
Duncan, Andrew Paul. "The analysis and application of artificial neural networks for early warning systems in hydrology and the environment." Thesis, University of Exeter, 2014. http://hdl.handle.net/10871/17569.
Abstract:
Artificial Neural Networks (ANNs) have been comprehensively researched, both from a computer scientific perspective and with regard to their use for predictive modelling in a wide variety of applications including hydrology and the environment. Yet their adoption for live, real-time systems remains on the whole sporadic and experimental. A plausible hypothesis is that this may be at least in part due to their treatment heretofore as “black boxes” that implicitly contain something that is unknown, or even unknowable. It is understandable that many of those responsible for delivering Early Warning Systems (EWS) might not wish to take the risk of implementing solutions perceived as containing unknown elements, despite the computational advantages that ANNs offer. This thesis therefore builds on existing efforts to open the box and develop tools and techniques that visualise, analyse and use ANN weights and biases especially from the viewpoint of neural pathways from inputs to outputs of feedforward networks. In so doing, it aims to demonstrate novel approaches to self-improving predictive model construction for both regression and classification problems. This includes Neural Pathway Strength Feature Selection (NPSFS), which uses ensembles of ANNs trained on differing subsets of data and analysis of the learnt weights to infer degrees of relevance of the input features and so build simplified models with reduced input feature sets. Case studies are carried out for prediction of flooding at multiple nodes in urban drainage networks located in three urban catchments in the UK, which demonstrate rapid, accurate prediction of flooding both for regression and classification. Predictive skill is shown to reduce beyond the time of concentration of each sewer node, when actual rainfall is used as input to the models. Further case studies model and predict statutory bacteria count exceedances for bathing water quality compliance at 5 beaches in Southwest England. An illustrative case study using a forest fires dataset from the UCI machine learning repository is also included. Results from these model ensembles generally exhibit improved performance, when compared with single ANN models. Also ensembles with reduced input feature sets, using NPSFS, demonstrate as good or improved performance when compared with the full feature set models. Conclusions are drawn about a new set of tools and techniques, including NPSFS and visualisation techniques for inspection of ANN weights, the adoption of which it is hoped may lead to improved confidence in the use of ANN for live real-time EWS applications.
30
Del, Vecchio Matteo. "Improving Deep Question Answering: The ALBERT Model." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/20414/.
Abstract:
Natural Language Processing is a field of Artificial Intelligence referring to the ability of computers to understand human speech and language, often in a written form, mainly by using Machine Learning and Deep Learning methods to extract patterns.
Languages are challenging by definition, because of their differences, their abstractions and their ambiguities; consequently, their processing is often very demanding, in terms of modelling the problem and resources. Retrieving all sentences in a given text is something that can be easily accomplished with just few lines of code, but what about checking whether a given sentence conveys a message with sarcasm or not? This is something difficult for humans too and therefore, it requires complex modelling mechanisms to be addressed. This kind of information, in fact, poses the problem of its encoding and representation in a meaningful way.
The majority of research involves finding and understanding all characteristics of text, in order to develop sophisticated models to address tasks such as Machine Translation, Text Summarization and Question Answering.
This work will focus on ALBERT, from Google Research, which is one of the recently released state-of-the-art models and investigate its performance on the Question Answering task. In addition, some ideas will be developed and experimented in order to improve model's performance on the Stanford Question Answering Dataset (SQuAD), after exploring breakthrough changes that made training and fine-tuning of huge language models possible.
31
Kišš, Martin. "Rozpoznávání historických textů pomocí hlubokých neuronových sítí." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2018. http://www.nusl.cz/ntk/nusl-385912.
Abstract:
The aim of this work is to create a tool for automatic transcription of historical documents. The work is mainly focused on the recognition of texts from the period of modern times written using font Fraktur. The problem is solved with a newly designed recurrent convolutional neural networks and a Spatial Transformer Network. Part of the solution is also an implemented generator of artificial historical texts. Using this generator, an artificial data set is created on which the convolutional neural network for line recognition is trained. This network is then tested on real historical lines of text on which the network achieves up to 89.0 % of character accuracy. The contribution of this work is primarily the newly designed neural network for text line recognition and the implemented artificial text generator, with which it is possible to train the neural network to recognize real historical lines of text.
32
Chen, Jianan. "Deep Learning Based Multimodal Retrieval." Electronic Thesis or Diss., Rennes, INSA, 2023. http://www.theses.fr/2023ISAR0019.
Abstract:
Les tâches multimodales jouent un rôle crucial dans la progression vers l'atteinte de l'intelligence artificielle (IA) générale. L'objectif principal de la recherche multimodale est d'exploiter des algorithmes d'apprentissage automatique pour extraire des informations sémantiques pertinentes, en comblant le fossé entre différentes modalités telles que les images visuelles, le texte linguistique et d'autres sources de données. Il convient de noter que l'entropie de l'information associée à des données hétérogènes pour des sémantiques de haut niveau identiques varie considérablement, ce qui pose un défi important pour les modèles multimodaux. Les modèles de réseau multimodal basés sur l'apprentissage profond offrent une solution efficace pour relever les difficultés découlant des différences substantielles d'entropie de l’information. Ces modèles présentent une précision et une stabilité impressionnantes dans les tâches d'appariement d'informations multimodales à grande échelle, comme la recherche d'images et de textes. De plus, ils démontrent de solides capacités d'apprentissage par transfert, permettant à un modèle bien entraîné sur une tâche multimodale d'être affiné et appliqué à une nouvelle tâche multimodale. Dans nos recherches, nous développons une nouvelle base de données multimodale et multi-vues générative spécifiquement conçue pour la tâche de segmentation référentielle multimodale. De plus, nous établissons une référence de pointe (SOTA) pour les modèles de segmentation d'expressions référentielles dans le domaine multimodal. Les résultats de nos expériences comparatives sont présentés de manière visuelle, offrant des informations claires et complètesMultimodal tasks play a crucial role in the progression towards achieving general artificial intelligence (AI). The primary goal of multimodal retrieval is to employ machine learning algorithms to extract relevant semantic information, bridging the gap between different modalities such as visual images, linguistic text, and other data sources. It is worth noting that the information entropy associated with heterogeneous data for the same high-level semantics varies significantly, posing a significant challenge for multimodal models. Deep learning-based multimodal network models provide an effective solution to tackle the difficulties arising from substantial differences in information entropy. These models exhibit impressive accuracy and stability in large-scale cross-modal information matching tasks, such as image-text retrieval. Furthermore, they demonstrate strong transfer learning capabilities, enabling a well-trained model from one multimodal task to be fine-tuned and applied to a new multimodal task, even in scenarios involving few-shot or zero-shot learning. In our research, we develop a novel generative multimodal multi-view database specifically designed for the multimodal referential segmentation task. Additionally, we establish a state-of-the-art (SOTA) benchmark and multi-view metric for referring expression segmentation models in the multimodal domain. The results of our comparative experiments are presented visually, providing clear and comprehensive insights
33
Bustos, Aurelia. "Extraction of medical knowledge from clinical reports and chest x-rays using machine learning techniques." Doctoral thesis, Universidad de Alicante, 2019. http://hdl.handle.net/10045/102193.
Abstract:
This thesis addresses the extraction of medical knowledge from clinical text using deep learning techniques. In particular, the proposed methods focus on cancer clinical trial protocols and chest x-rays reports. The main results are a proof of concept of the capability of machine learning methods to discern which are regarded as inclusion or exclusion criteria in short free-text clinical notes, and a large scale chest x-ray image dataset labeled with radiological findings, diagnoses and anatomic locations. Clinical trials provide the evidence needed to determine the safety and effectiveness of new medical treatments. These trials are the basis employed for clinical practice guidelines and greatly assist clinicians in their daily practice when making decisions regarding treatment. However, the eligibility criteria used in oncology trials are too restrictive. Patients are often excluded on the basis of comorbidity, past or concomitant treatments and the fact they are over a certain age, and those patients that are selected do not, therefore, mimic clinical practice. This signifies that the results obtained in clinical trials cannot be extrapolated to patients if their clinical profiles were excluded from the clinical trial protocols. The efficacy and safety of new treatments for patients with these characteristics are not, therefore, defined. Given the clinical characteristics of particular patients, their type of cancer and the intended treatment, discovering whether or not they are represented in the corpus of available clinical trials requires the manual review of numerous eligibility criteria, which is impracticable for clinicians on a daily basis. In this thesis, a large medical corpora comprising all cancer clinical trials protocols in the last 18 years published by competent authorities was used to extract medical knowledge in order to help automatically learn patient’s eligibility in these trials. For this, a model is built to automatically predict whether short clinical statements were considered inclusion or exclusion criteria. A method based on deep neural networks is trained on a dataset of 6 million short free-texts to classify them between elegible or not elegible. For this, pretrained word embeddings were used as inputs in order to predict whether or not short free-text statements describing clinical information were considered eligible. The semantic reasoning of the word-embedding representations obtained was also analyzed, being able to identify equivalent treatments for a type of tumor in an analogy with the drugs used to treat other tumors. Results show that representation learning using deep neural networks can be successfully leveraged to extract the medical knowledge from clinical trial protocols and potentially assist practitioners when prescribing treatments. The second main task addressed in this thesis is related to knowledge extraction from medical reports associated with radiographs. Conventional radiology remains the most performed technique in radiodiagnosis services, with a percentage close to 75% (Radiología Médica, 2010). In particular, chest x-ray is the most common medical imaging exam with over 35 million taken every year in the US alone (Kamel et al., 2017). They allow for inexpensive screening of several pathologies including masses, pulmonary nodules, effusions, cardiac abnormalities and pneumothorax. For this task, all the chest-x rays that had been interpreted and reported by radiologists at the Hospital Universitario de San Juan (Alicante) from Jan 2009 to Dec 2017 were used to build a novel large-scale dataset in which each high-resolution radiograph is labeled with its corresponding metadata, radiological findings and pathologies. This dataset, named PadChest, includes more than 160,000 images obtained from 67,000 patients, covering six different position views and additional information on image acquisition and patient demography. The free text reports written in Spanish by radiologists were labeled with 174 different radiographic findings, 19 differential diagnoses and 104 anatomic locations organized as a hierarchical taxonomy and mapped onto standard Unified Medical Language System (UMLS) terminology. For this, a subset of the reports (a 27%) were manually annotated by trained physicians, whereas the remaining set was automatically labeled with deep supervised learning methods using attention mechanisms and fed with the text reports. The labels generated were then validated in an independent test set achieving a 0.93 Micro-F1 score. To the best of our knowledge, this is one of the largest public chest x-ray databases suitable for training supervised models concerning radiographs, and also the first to contain radiographic reports in Spanish. The PadChest dataset can be downloaded on request from http://bimcv.cipf.es/bimcv-projects/padchest/. PadChest is intended for training image classifiers based on deep learning techniques to extract medical knowledge from chest x-rays. It is essential that automatic radiology reporting methods could be integrated in a clinically validated manner in radiologists’ workflow in order to help specialists to improve their efficiency and enable safer and actionable reporting. Computer vision methods capable of identifying both the large spectrum of thoracic abnormalities (and also the normality) need to be trained on large-scale comprehensively labeled large-scale x-ray datasets such as PadChest. The development of these computer vision tools, once clinically validated, could serve to fulfill a broad range of unmet needs. Beyond implementing and obtaining results for both clinical trials and chest x-rays, this thesis studies the nature of the health data, the novelty of applying deep learning methods to obtain large-scale labeled medical datasets, and the relevance of its applications in medical research, which have contributed to its extramural diffusion and worldwide reach. This thesis describes this journey so that the reader is navigated across multiple disciplines, from engineering to medicine up to ethical considerations in artificial intelligence applied to medicine.
34
Alsulami, Khalil Ibrahim D. "Application-Based Network Traffic Generator for Networking AI Model Development." University of Dayton / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1619387614152354.
35
Štarha, Dominik. "Meření podobnosti obrazů s pomocí hlubokého učení." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2018. http://www.nusl.cz/ntk/nusl-377018.
Abstract:
This master´s thesis deals with the reseach of technologies using deep learning method, being able to use when processing image data. Specific focus of the work is to evaluate the suitability and effectiveness of deep learning when comparing two image input data. The first – theoretical – part consists of the introduction to neural networks and deep learning. Also, it contains a description of available methods, their benefits and principles, used for processing image data. The second - practical - part of the thesis contains a proposal a appropriate model of Siamese networks to solve the problem of comparing two input image data and evaluating their similarity. The output of this work is an evaluation of several possible model configurations and highlighting the best-performing model parameters.
36
Varga, Adam. "Identifikace a charakterizace škodlivého chování v grafech chování." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2021. http://www.nusl.cz/ntk/nusl-442388.
Abstract:
Za posledné roky je zaznamenaný nárast prác zahrňujúcich komplexnú detekciu malvéru. Pre potreby zachytenia správania je často vhodné pouziť formát grafov. To je prípad antivírusového programu Avast, ktorého behaviorálny štít deteguje škodlivé správanie a ukladá ich vo forme grafov. Keďže sa jedná o proprietárne riešenie a Avast antivirus pracuje s vlastnou sadou charakterizovaného správania bolo nutné navrhnúť vlastnú metódu detekcie, ktorá bude postavená nad týmito grafmi správania. Táto práca analyzuje grafy správania škodlivého softvéru zachytené behavioralnym štítom antivírusového programu Avast pre proces hlbšej detekcie škodlivého softvéru. Detekcia škodlivého správania sa začína analýzou a abstrakciou vzorcov z grafu správania. Izolované vzory môžu efektívnejšie identifikovať dynamicky sa meniaci malware. Grafy správania sú uložené v databáze grafov Neo4j a každý deň sú zachytené tisíce z nich. Cieľom tejto práce bolo navrhnúť algoritmus na identifikáciu správania škodlivého softvéru s dôrazom na rýchlosť skenovania a jasnosť identifikovaných vzorcov správania. Identifikácia škodlivého správania spočíva v nájdení najdôležitejších vlastností natrénovaných klasifikátorov a následnej extrakcie podgrafu pozostávajúceho iba z týchto dôležitých vlastností uzlov a vzťahov medzi nimi. Následne je navrhnuté pravidlo pre hodnotenie extrahovaného podgrafu. Diplomová práca prebehla v spolupráci so spoločnosťou Avast Software s.r.o.
37
Wåhlin, Peter. "Enhanching the Human-Team Awareness of a Robot." Thesis, Mälardalens högskola, Akademin för innovation, design och teknik, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:mdh:diva-16371.
Abstract:
The use of autonomous robots in our society is increasing every day and a robot is no longer seen as a tool but as a team member. The robots are now working side by side with us and provide assistance during dangerous operations where humans otherwise are at risk. This development has in turn increased the need of robots with more human-awareness. Therefore, this master thesis aims at contributing to the enhancement of human-aware robotics. Specifically, we are investigating the possibilities of equipping autonomous robots with the capability of assessing and detecting activities in human teams. This capability could, for instance, be used in the robot's reasoning and planning components to create better plans that ultimately would result in improved human-robot teamwork performance. we propose to improve existing teamwork activity recognizers by adding intangible features, such as stress, motivation and focus, originating from human behavior models. Hidden markov models have earlier been proven very efficient for activity recognition and have therefore been utilized in this work as a method for classification of behaviors. In order for a robot to provide effective assistance to a human team it must not only consider spatio-temporal parameters for team members but also the psychological.To assess psychological parameters this master thesis suggests to use the body signals of team members. Body signals such as heart rate and skin conductance. Combined with the body signals we investigate the possibility of using System Dynamics models to interpret the current psychological states of the human team members, thus enhancing the human-awareness of a robot.Användningen av autonoma robotar i vårt samhälle ökar varje dag och en robot ses inte längre som ett verktyg utan som en gruppmedlem. Robotarna arbetar nu sida vid sida med oss och ger oss stöd under farliga arbeten där människor annars är utsatta för risker. Denna utveckling har i sin tur ökat behovet av robotar med mer människo-medvetenhet. Därför är målet med detta examensarbete att bidra till en stärkt människo-medvetenhet hos robotar. Specifikt undersöker vi möjligheterna att utrusta autonoma robotar med förmågan att bedöma och upptäcka olika beteenden hos mänskliga lag. Denna förmåga skulle till exempel kunna användas i robotens resonemang och planering för att ta beslut och i sin tur förbättra samarbetet mellan människa och robot. Vi föreslår att förbättra befintliga aktivitetsidentifierare genom att tillföra förmågan att tolka immateriella beteenden hos människan, såsom stress, motivation och fokus. Att kunna urskilja lagaktiviteter inom ett mänskligt lag är grundläggande för en robot som ska vara till stöd för laget. Dolda markovmodeller har tidigare visat sig vara mycket effektiva för just aktivitetsidentifiering och har därför använts i detta arbete. För att en robot ska kunna ha möjlighet att ge ett effektivt stöd till ett mänskligtlag måste den inte bara ta hänsyn till rumsliga parametrar hos lagmedlemmarna utan även de psykologiska. För att tyda psykologiska parametrar hos människor förespråkar denna masteravhandling utnyttjandet av mänskliga kroppssignaler. Signaler så som hjärtfrekvens och hudkonduktans. Kombinerat med kroppenssignalerar påvisar vi möjligheten att använda systemdynamiksmodeller för att tolka immateriella beteenden, vilket i sin tur kan stärka människo-medvetenheten hos en robot.
The thesis work was conducted in Stockholm, Kista at the department of Informatics and Aero System at Swedish Defence Research Agency.
38
"Anomaly Detection in Categorical Datasets with Artificial Contrasts." Master's thesis, 2016. http://hdl.handle.net/2286/R.I.40782.
Abstract:
abstract: Anomaly is a deviation from the normal behavior of the system and anomaly detection techniques try to identify unusual instances based on deviation from the normal data. In this work, I propose a machine-learning algorithm, referred to as Artificial Contrasts, for anomaly detection in categorical data in which neither the dimension, the specific attributes involved, nor the form of the pattern is known a priori. I use RandomForest (RF) technique as an effective learner for artificial contrast. RF is a powerful algorithm that can handle relations of attributes in high dimensional data and detect anomalies while providing probability estimates for risk decisions.
I apply the model to two simulated data sets and one real data set. The model was able to detect anomalies with a very high accuracy. Finally, by comparing the proposed model with other models in the literature, I demonstrate superior performance of the proposed model.Dissertation/Thesis
Masters Thesis Industrial Engineering 2016
39
"Understanding the Importance of Entities and Roles in Natural Language Inference : A Model and Datasets." Master's thesis, 2019. http://hdl.handle.net/2286/R.I.54921.
Abstract:
abstract: In this thesis, I present two new datasets and a modification to the existing models in the form of a novel attention mechanism for Natural Language Inference (NLI). The new datasets have been carefully synthesized from various existing corpora released for different tasks.
The task of NLI is to determine the possibility of a sentence referred to as “Hypothesis” being true given that another sentence referred to as “Premise” is true. In other words, the task is to identify whether the “Premise” entails, contradicts or remains neutral with regards to the “Hypothesis”. NLI is a precursor to solving many Natural Language Processing (NLP) tasks such as Question Answering and Semantic Search. For example, in Question Answering systems, the question is paraphrased to form a declarative statement which is treated as the hypothesis. The options are treated as the premise. The option with the maximum entailment score is considered as the answer. Considering the applications of NLI, the importance of having a strong NLI system can't be stressed enough.
Many large-scale datasets and models have been released in order to advance the field of NLI. While all of these models do get good accuracy on the test sets of the datasets they were trained on, they fail to capture the basic understanding of “Entities” and “Roles”. They often make the mistake of inferring that “John went to the market.” from “Peter went to the market.” failing to capture the notion of “Entities”. In other cases, these models don't understand the difference in the “Roles” played by the same entities in “Premise” and “Hypothesis” sentences and end up wrongly inferring that “Peter drove John to the stadium.” from “John drove Peter to the stadium.”
The lack of understanding of “Roles” can be attributed to the lack of such examples in the various existing datasets. The reason for the existing model’s failure in capturing the notion of “Entities” is not just due to the lack of such examples in the existing NLI datasets. It can also be attributed to the strict use of vector similarity in the “word-to-word” attention mechanism being used in the existing architectures.
To overcome these issues, I present two new datasets to help make the NLI systems capture the notion of “Entities” and “Roles”. The “NER Changed” (NC) dataset and the “Role-Switched” (RS) dataset contains examples of Premise-Hypothesis pairs that require the understanding of “Entities” and “Roles” respectively in order to be able to make correct inferences. This work shows how the existing architectures perform poorly on the “NER Changed” (NC) dataset even after being trained on the new datasets. In order to help the existing architectures, understand the notion of “Entities”, this work proposes a modification to the “word-to-word” attention mechanism. Instead of relying on vector similarity alone, the modified architectures learn to incorporate the “Symbolic Similarity” as well by using the Named-Entity features of the Premise and Hypothesis sentences. The new modified architectures not only perform significantly better than the unmodified architectures on the “NER Changed” (NC) dataset but also performs as well on the existing datasets.Dissertation/Thesis
Masters Thesis Computer Science 2019
40
Lee, Chia-Yi, and 李佳怡. "Investigating the Hybrid Models of Decision Tree, Logistic Regression and Artificial Neural Network for Predicting Recurrence of Breast Cancer from Public Microarray Datasets." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/72125064198285323874.
41
Mendonça, Francisco de Andrade Bravo. "AudioMood: Classificação de emoções em bandas sonoras de filmes usando Redes Neuronais." Master's thesis, 2021. http://hdl.handle.net/10451/49347.
Abstract:
Tese de mestrado, Informática, Universidade de Lisboa, Faculdade de Ciências, 2021O recurso à Inteligência Artificial para a ajuda ou execução de uma tarefa é cada vez mais frequente na nossa vida. Desde assistentes pessoais e médicos ou até carros autónomos, o uso é vasto e é adoptado nas mais diversas áreas. Com o aumentar de complexidade das AI, estas requerem a criação de novos métodos para melhorar o treino de tarefas complexas. Nesse sentido, esta dissertação tenta ajudar o estudo dos métodos de treino de Redes Neuronais, utilizando áudio de modo a que a rede consiga identificar os sons presentes num filme. Para concretizar esse objectivo, o primeiro passo foi a análise de diversos datasets, de forma a seleccionar um que seja adaptado à metodologia utilizada. O dataset escolhido foi o AudioSet da Google, pois tem mais de dois milhões de vídeos anotados, algo que favorece este estudo. De seguida, foram desenvolvidas ferramentas para a criação de conjuntos mais pequenos de dados com base no AudioSet. Estas ferramentas trataram do download dos vídeos, a sua conversão em áudio, a manipulação e tratamento dos últimos, e a construção de novos datasets. No processo anteriormente descrito, foram aplicados os métodos de aumentação de dados, sendo estes a rotação de dados e o controlo de volume. Após a criação do dataset procedeu-se o treino. Para cada treino foi utilizado a mesma arquitectura do modelo, com pequenas diferenças no método de treino. É possível afirmar que para a tarefa escolhida, o aumento de dados no dataset e o uso de rotação de dados melhorou os resultados, enquanto a manipulação de volumes não ofereceu alterações suficientes aos dados para permitir que o modelo melhorasse.
Nowadays the use of Artificial Intelligence to help or execute a task is ever more frequent. From personal assistants, to video games, to autonomous cars, the ability to use AI is vast, and getting adopted in new areas. As the complexity of AI increases, the necessity of developing new methods to help in the training of AI is critical. In that sense, this dissertation tries to help in the study training methods for Neural Networks, using audio sources, so that it is able to identify the different sounds present in a movie. To meet this purpose, the first step was the analysis of different datasets, to find one that is adaptable to the methodology used. The chosen dataset was AudioSet by Google, which has more than 2 million annotated videos. Later, tools were developed to create smaller datasets from AudioSet. These tools took care of video download, their conversion to audio, the manipulation and treatment of these audios, and the construction of new datasets. In this process, data rotation and volume control, two methods of data augmentation, were applied with the intention of creating new data. With the abovementioned new dataset, models were trained. The same model architecture was used for all the training processes, but with small differences in the training method. For the chosen task, it can be said that the increase of data in the dataset and the use of data rotation improved the test results, while volume control didn’t offer sufficient alterations to the data, and so the test results didn’t improve.
42
(8768079), Nanxin Jin. "ASD PREDICTION FROM STRUCTURAL MRI WITH MACHINE LEARNING." Thesis, 2020.
Abstract:
Autism Spectrum Disorder (ASD) is part of the developmental disabilities. There are numerous symptoms for ASD patients, including lack of abilities in social interaction, communication obstacle and repeatable behaviors. Meanwhile, the rate of ASD prevalence has kept rising by the past 20 years from 1 out of 150 in 2000 to 1 out of 54 in 2016. In addition, the ASD population is quite large. Specifically, 3.5 million Americans live with ASD in the year of 2014, which will cost U.S. citizens $236-$262 billion dollars annually for autism services. So, it is critical to make an accurate diagnosis for preschool age children with ASD, in order to give them a better life. Instead of using traditional ASD behavioral tests, such as ADI-R, ADOS, and DSM-IV, we applied brain MRI images as input to make diagnosis. We revised 3D-ResNet structure to fit 110 preschool children's brain MRI data, along with Convolution 3D and VGG model. The prediction accuracy with raw data is 65.22%. The accuracy is significantly improved to 82.61% by removing the noise around the brain. We also showed the speed of ML prediction is 308 times faster than behavior tests.
43
"Referring Expression Comprehension for CLEVR-Ref+ Dataset." Master's thesis, 2020. http://hdl.handle.net/2286/R.I.62696.
Abstract:
abstract: Referring Expression Comprehension (REC) is an important area of research in Natural Language Processing (NLP) and vision domain. It involves locating an object in an image described by a natural language referring expression. This task requires information from both Natural Language and Vision aspect. The task is compositional in nature as it requires visual reasoning as underlying process along with relationships among the objects in the image. Recent works based on modular networks have
displayed to be an effective framework for performing visual reasoning task.
Although this approach is effective, it has been established that the current benchmark datasets for referring expression comprehension suffer from bias. Recent work on CLEVR-Ref+ dataset deals with bias issues by constructing a synthetic dataset
and provides an approach for the aforementioned task which performed better than the previous state-of-the-art models as well as showing the reasoning process. This work aims to improve the performance on CLEVR-Ref+ dataset and achieve comparable interpretability. In this work, the neural module network approach with the attention map technique is employed. The neural module network is composed of the primitive operation modules which are specific to their functions and the output is generated using a separate segmentation module. From empirical results, it is clear that this approach is performing significantly better than the current State-of-theart in one aspect (Predicted programs) and achieving comparable results for another aspect (Ground truth programs)Dissertation/Thesis
Masters Thesis Computer Science 2020
44
Gandhi, Priyanka. "Extracting Symptoms from Narrative Text using Artificial Intelligence." Thesis, 2020. http://hdl.handle.net/1805/24759.
Abstract:
Indiana University-Purdue University Indianapolis (IUPUI)Electronic health records collect an enormous amount of data about patients. However, the information about the patient’s illness is stored in progress notes that are in an un- structured format. It is difficult for humans to annotate symptoms listed in the free text. Recently, researchers have explored the advancements of deep learning can be applied to pro- cess biomedical data. The information in the text can be extracted with the help of natural language processing. The research presented in this thesis aims at automating the process of symptom extraction. The proposed methods use pre-trained word embeddings such as BioWord2Vec, BERT, and BioBERT to generate vectors of the words based on semantics and syntactic structure of sentences. BioWord2Vec embeddings are fed into a BiLSTM neural network with a CRF layer to capture the dependencies between the co-related terms in the sentence. The pre-trained BERT and BioBERT embeddings are fed into the BERT model with a CRF layer to analyze the output tags of neighboring tokens. The research shows that with the help of the CRF layer in neural network models, longer phrases of symptoms can be extracted from the text. The proposed models are compared with the UMLS Metamap tool that uses various sources to categorize the terms in the text to different semantic types and Stanford CoreNLP, a dependency parser, that analyses syntactic relations in the sentence to extract information. The performance of the models is analyzed by using strict, relaxed, and n-gram evaluation schemes. The results show BioBERT with a CRF layer can extract the majority of the human-labeled symptoms. Furthermore, the model is used to extract symptoms from COVID-19 tweets. The model was able to extract symptoms listed by CDC as well as new symptoms.
45
Lima, Bruno Tiago da Silva. "VISUALIZAÇÃO E ANÁLISE EM OBSERVAÇÕES AÉREAS POR INTELIGÊNCIA ARTIFICIAL." Master's thesis, 2020. http://hdl.handle.net/11110/1965.
Abstract:
Atualmente as imagens de observação aérea permitem visualizar o mundo numa outra perspetiva, possuindo cada vez mais importância em estudos e planeamentos como, por exemplo, planos e ordenamentos de território, mapas de danos em caso de catástrofes naturais, medição de áreas de desflorestação, entre outros.
Os Municípios têm como principal responsabilidade a gestão administrativa do seu território e, para isso, recorrem a serviços de informação geográfica e ortofotomapas para apoio a essa gestão. Uma das grandes dificuldades prende-se com a rápida mudança que acontece no território, sendo difícil a fiscalização dessas mesmas alterações. Partindo da necessidade dos Municípios na obtenção de um cadastro atualizado das piscinas existentes no concelho, de forma a perceber as mudanças no território e validar o licenciamento, neste projeto procurou-se desenvolver um protótipo de sistema, utilizando e integrando um conjunto de tecnologias que incluiu inteligência artificial, middleware para integração de sistemas, sistemas de informação geográfica e bibliotecas de visualização avançada. A inteligência artificial contribuiu para a deteção de determinadas construções, identificadas como piscinas; um middleware de serviços permitiu integrar o resultado dessa identificação com o sistema de informação geográfica do Município para georreferenciar as mesmas, cruzando essa informação com os dados relativos ao licenciamento de obras particulares. Com as bibliotecas de visualização foi criada uma interface avançada de apoio ao utilizador para a visualização e análise da informação.
Para efetuar a deteção das piscinas foi necessário criar um dataset, constituído por imagens aéreas de piscinas, e efetuar a anotação da região relevante na qual é identificada a piscina.
Foi treinado o sistema em várias redes neuronais convolucionais para a deteção das piscinas, de forma a obter e comparar a precisão dos resultados obtidos.
46
"Accessible Retail Shopping For The Visually Impaired Using Deep Learning." Master's thesis, 2020. http://hdl.handle.net/2286/R.I.57075.
Abstract:
abstract: Over the past decade, advancements in neural networks have been instrumental in achieving remarkable breakthroughs in the field of computer vision. One of the applications is in creating assistive technology to improve the lives of visually impaired people by making the world around them more accessible. A lot of research in convolutional neural networks has led to human-level performance in different vision tasks including image classification, object detection, instance segmentation, semantic segmentation, panoptic segmentation and scene text recognition. All the before mentioned tasks, individually or in combination, have been used to create assistive technologies to improve accessibility for the blind.
This dissertation outlines various applications to improve accessibility and independence for visually impaired people during shopping by helping them identify products in retail stores. The dissertation includes the following contributions; (i) A dataset containing images of breakfast-cereal products and a classifier using a deep neural (ResNet) network; (ii) A dataset for training a text detection and scene-text recognition model; (iii) A model for text detection and scene-text recognition to identify product images using a user-controlled camera; (iv) A dataset of twenty thousand products with product information and related images that can be used to train and test a system designed to identify products.Dissertation/Thesis
Masters Thesis Computer Science 2020
47
Foroozandeh, Mehdi. "GAN-Based Synthesis of Brain Tumor Segmentation Data : Augmenting a dataset by generating artificial images." Thesis, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-169863.
Abstract:
Machine learning applications within medical imaging often suffer from a lack of data, as a consequence of restrictions that hinder the free distribution of patient information. In this project, GANs (generative adversarial networks) are used to generate data synthetically, in an effort to circumvent this issue. The GAN framework PGAN is trained on the brain tumor segmentation dataset BraTS to generate new, synthetic brain tumor masks with the same visual characteristics as the real samples. The image-to-image translation network SPADE is subsequently trained on the image pairs in the real dataset, to learn a transformation from segmentation masks to brain MR images, and is in turn used to map the artificial segmentation masks generated by PGAN to corresponding artificial MR images. The images generated by these networks form a new, synthetic dataset, which is used to augment the original dataset. Different quantities of real and synthetic data are then evaluated in three different brain tumor segmentation tasks, where the image segmentation network U-Net is trained on this data to segment (real) MR images into the classes in question. The final segmentation performance of each training instance is evaluated over test data from the real dataset with the Weighted Dice Loss metric. The results indicate a slight increase in performance across all segmentation tasks evaluated in this project, when including some quantity of synthetic images. However, the differences were largest when the experiments were restricted to using only 20 % of the real data, and less significant when the full dataset was made available. A majority of the generated segmentation masks appear visually convincing to an extent (although somewhat noisy with regards to the intra-tumoral classes), while a relatively large proportion appear heavily noisy and corrupted. However, the translation of segmentation masks to MR images via SPADE proved more reliable and consistent.
48
(5929832), Ikbeom Jang. "Diffusion Tensor Imaging Analysis for Subconcussive Trauma in Football and Convolutional Neural Network-Based Image Quality Control That Does Not Require a Big Dataset." Thesis, 2019.
Abstract:
Diffusion Tensor Imaging (DTI) is a magnetic resonance imaging (MRI)-based technique that has frequently been used for the identification of brain biomarkers of neurodevelopmental and neurodegenerative disorders because of its ability to assess the structural organization of brain tissue. In this work, I present (1) preclinical findings of a longitudinal DTI study that investigated asymptomatic high school football athletes who experienced repetitive head impact and (2) an automated pipeline for assessing the quality of DTI images that uses a convolutional neural network (CNN) and transfer learning. The first section addresses the effects of repetitive subconcussive head trauma on the white matter of adolescent brains. Significant concerns exist regarding sub-concussive injury in football since many studies have reported that repetitive blows to the head may change the microstructure of white matter. This is more problematic in youth-aged athletes whose white matter is still developing. Using DTI and head impact monitoring sensors, regions of significantly altered white matter were identified and within-season effects of impact exposure were characterized by identifying the volume of regions showing significant changes for each individual. The second section presents a novel pipeline for DTI quality control (QC). The complex nature and long acquisition time associated with DTI make it susceptible to artifacts that often result in inferior diagnostic image quality. We propose an automated QC algorithm based on a deep convolutional neural network (DCNN). Adaptation of transfer learning makes it possible to train a DCNN with a relatively small dataset in a short time. The QA algorithm detects not only motion- or gradient-related artifacts, but also various erroneous acquisitions, including images with regional signal loss or those that have been incorrectly imaged or reconstructed.
49
Dale, Ashley S. "3D Object Detection Using Virtual Environment Assisted Deep Network Training." Thesis, 2020. http://hdl.handle.net/1805/24756.
Abstract:
Indiana University-Purdue University Indianapolis (IUPUI)An RGBZ synthetic dataset consisting of five object classes in a variety of virtual environments and orientations was combined with a small sample of real-world image data and used to train the Mask R-CNN (MR-CNN) architecture in a variety of configurations. When the MR-CNN architecture was initialized with MS COCO weights and the heads were trained with a mix of synthetic data and real world data, F1 scores improved in four of the five classes: The average maximum F1-score of all classes and all epochs for the networks trained with synthetic data is F1∗ = 0.91, compared to F1 = 0.89 for the networks trained exclusively with real data, and the standard deviation of the maximum mean F1-score for synthetically trained networks is σ∗ = 0.015, compared to σ_F1 = 0.020 for the networks trained exclusively with real F1 data. Various backgrounds in synthetic data were shown to have negligible impact on F1 scores, opening the door to abstract backgrounds and minimizing the need for intensive synthetic data fabrication. When the MR-CNN architecture was initialized with MS COCO weights and depth data was included in the training data, the net- work was shown to rely heavily on the initial convolutional input to feed features into the network, the image depth channel was shown to influence mask generation, and the image color channels were shown to influence object classification. A set of latent variables for a subset of the synthetic datatset was generated with a Variational Autoencoder then analyzed using Principle Component Analysis and Uniform Manifold Projection and Approximation (UMAP). The UMAP analysis showed no meaningful distinction between real-world and synthetic data, and a small bias towards clustering based on image background.
50
(8771429), Ashley S. Dale. "3D OBJECT DETECTION USING VIRTUAL ENVIRONMENT ASSISTED DEEP NETWORK TRAINING." Thesis, 2021.
Abstract:
An RGBZ synthetic dataset consisting of five object classes in a variety of virtual environments and orientations was combined with a small sample of real-world image data and used to train the Mask R-CNN (MR-CNN) architecture in a variety of configurations. When the MR-CNN architecture was initialized with MS COCO weights and the heads were trained with a mix of synthetic data and real world data, F1 scores improved in four of the five classes: The average maximum F1-score of all classes and all epochs for the networks trained with synthetic data is F1∗ = 0.91, compared to F1 = 0.89 for the networks trained exclusively with real data, and the standard deviation of the maximum mean F1-score for synthetically trained networks is σ∗ F1 = 0.015, compared to σF 1 = 0.020 for the networks trained exclusively with real data. Various backgrounds in synthetic data were shown to have negligible impact on F1 scores, opening the door to abstract backgrounds and minimizing the need for intensive synthetic data fabrication. When the MR-CNN architecture was initialized with MS COCO weights and depth data was included in the training data, the net- work was shown to rely heavily on the initial convolutional input to feed features into the network, the image depth channel was shown to influence mask generation, and the image color channels were shown to influence object classification. A set of latent variables for a subset of the synthetic datatset was generated with a Variational Autoencoder then analyzed using Principle Component Analysis and Uniform Manifold Projection and Approximation (UMAP). The UMAP analysis showed no meaningful distinction between real-world and synthetic data, and a small bias towards clustering based on image background.