Dissertations / Theses on the topic 'Convolutional transformer'
To see the other types of publications on this topic, follow the link: Convolutional transformer.
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 'Convolutional transformer.'
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
Dronzeková, Michaela. "Analýza polygonálních modelů pomocí neuronových sítí." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2020. http://www.nusl.cz/ntk/nusl-417253.
Full textAbstract:
This thesis deals with rotation estimation of 3D model of human jaw. It describes and compares methods for direct analysis od 3D models as well as method to analyze model using rasterization. To evaluate perfomance of proposed method, a metric that computes number of cases when prediction was less than 30° from ground truth is used. Proposed method that uses rasterization, takes three x-ray views of model as an input and processes it with convolutional network. It achieves best preformance, 99% with described metric. Method to directly analyze polygonal model as a sequence uses attention mechanism to do so and was inspired by transformer architecture. A special pooling function was proposed for this network that decreases memory requirements of the network. This method achieves 88%, but does not use rasterization and can process polygonal model directly. It is not as good as rasterization method with x-ray display, byt it is better than rasterization method with model not rendered as x-ray. The last method uses graph representation of mesh. Graph network had problems with overfitting, that is why it did not get good results and I think this method is not very suitable for analyzing plygonal model.
2
Domini, Davide. "Classificazione di Radiografie Toraciche per la Diagnosi del COVID-19 con Reti Convoluzionali e Vision Transformer." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/24243/.
Full textAbstract:
Negli ultimi due anni, per via della pandemia generata dal virus Covid19, la vita in ogni angolo del nostro pianeta è drasticamente cambiata. Ad oggi, nel mondo, sono oltre duecentoventi milioni le persone che hanno contratto questo virus e sono quasi cinque milioni le persone decedute. In alcuni periodi si è arrivati ad avere anche un milione di nuovi contagiati al giorno e mediamente, negli ultimi sei mesi, questo dato è stato di più di mezzo milione al giorno.
Gli ospedali, soprattutto nei paesi meno sviluppati, hanno subito un grande stress e molte volte hanno avuto una carenza di risorse per fronteggiare questa grave pandemia. Per questo motivo ogni ricerca in questo campo diventa estremamente importante, soprattutto quelle che, con l'ausilio dell'intelligenza artificiale, riescono a dare supporto ai medici. Queste tecnologie una volta sviluppate e approvate possono essere diffuse a costi molto bassi e accessibili a tutti.
In questo elaborato sono stati sperimentati e valutati due diversi approcci alla diagnosi del Covid-19 a partire dalle radiografie toraciche dei pazienti: il primo metodo si basa sul transfer learning di una rete convoluzionale inizialmente pensata per la classificazione di immagini. Il secondo approccio utilizza i Vision Transformer (ViT), un'architettura ampiamente diffusa nel campo del Natural Language Processing adattata ai task di Visione Artificiale. La prima soluzione ha ottenuto un’accuratezza di 0.85 mentre la seconda di 0.92, questi risultati, soprattutto il secondo, sono molto incoraggianti soprattutto vista la minima quantità di dati di training necessaria.
3
Richtarik, Lukáš. "Rozpoznávání ručně psaného textu pomocí hlubokých neuronových sítí." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2020. http://www.nusl.cz/ntk/nusl-433517.
Full textAbstract:
The work deals with the issue of handrwritten text recognition problem with deep neural networks. It focuses on the use of sequence to sequence method using encoder-decoder model. It also includes design of encoder-decoder model for handwritten text recognition using a transformer instead of recurrent neurons and a set of experiments that were performed on it.
4
Duhamel, Pierre. "Algorithmes de transformées discrètes rapides pour convolution cyclique et de convolution cyclique pour transformées rapides." Paris 11, 1986. http://www.theses.fr/1986PA112197.
Full textAbstract:
Nous présentons tout d'abord les différentes transformées qui seront considérées dans ce travail (Transformées en Nombres Entiers - TNE -, Transformées de Fourier - TFR -, Transformées polynomiales - TP -, Transformées en Cosinus Discrètes - TCD -) de manière homogène. Puis, nous utilisons le lien entre TNE et TP pour proposer une nouvelle famille de TNE avec 2 comme racine de l'unité incluant des transformées classiques (Fermat, Mersenne, etc. . . ), mais aussi de nouvelles transformées de plus grande longueur pour une dynamique donnée. Nous montrons également une propriété de décomposition de l'arithmétique dans cette classe de transformées. Nous proposons également un ensemble d'algorithmes de TNE à nombre minimum de décalages. Dans le cadre des TFR, nous avons proposé un algorithme connu sous le nom de "split-radix". L'application de cet algorithme à des données complexes, réelles, ou réelles et symétriques a permis d'obtenir les programmes demandant à chaque fois les nombres d'opérations (multiplications et additions) les plus faibles connus, tout en gardant une structure régulière. Nous avons pu montrer que toute amélioration éventuelle de l'un de ces algorithmes se traduirait par une amélioration correspondante sur tous les autres. D'autre part, nous avons montré que leur structure était semblable à celle des algorithmes optimaux vis à vis du nombre de multiplications. La recherche d'algorithmes de TCD à nombre minimum de multiplications nous a permis de mettre en évidence l'équivalence entre une TCD de longueur 2n et une convolution cyclique. Ceci nous a conduits à proposer deux nouvelles architectures de TCD : l'une présente l'avantage de ne nécessiter qu'une multiplication générale par point de calcul, mais nécessite un arithmétique modulo. L'autre basée sur l'utilisation de l'arithmétique distribuée, présente une structure très simple, réutilisable pour d'autres types de transformées rapidesFirst, we present the fast transforms to be considered in the following (Number Theoretic Transforms - NTT -, Fast Fourier Transforms - FFT -, Polynomial Transforms - PT -, Discrete Cosine Transforms - OCT -) in a unified manner. Then, we use the link between NTT and PT to propose a new family of NTT with 2 as a root of unity which includes classical ones (Fermat, Mersenne, etc. . . ), but also includes new ones, which allow transforming longer sequences for a given dynamic range. We also establish a decomposition property of the arithmetic in this family of NTT's. We also propose a set of NTT algorithms with minimum number of shifts. Next, for the computation of FFT's, we introduced the "split-radix" algorithm. Application of this algorithm to complex, real or real-symmetric data requires in each case the smallest known number of operations (multiplications and additions), with a regular structure. We also showed that any improvement of one of these algorithms would also bring a corresponding improvement of the other ones. Furthermore, we showed that their structure is very similar to that of the optimum (minimum number of multiplications) algorithm. The search for minimum multiplication DCT algorithms enabled us to show the equivalence between a 2n DCT and a cyclic convolution. We used this equivalence to propose two new architectures for DCT computation. The first one presents the advantage of needing only one general multiplier per point, but necessitates modulo-arithmetic. The second one, based on distributed arithmetic, has a very simple structure that can be used also for other fast transforms (FFT, e. G. )
5
Kwok, Yien Chian. "Shah convolution Fourier transform detection." Thesis, Imperial College London, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.428105.
Full text
6
Westermark, Pontus. "Wavelets, Scattering transforms and Convolutional neural networks : Tools for image processing." Thesis, Uppsala universitet, Analys och sannolikhetsteori, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-337570.
Full text
7
Highlander, Tyler. "Efficient Training of Small Kernel Convolutional Neural Networks using Fast Fourier Transform." Wright State University / OhioLINK, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=wright1432747175.
Full text
8
Martucci, Stephen A. "Symmetric convolution and the discrete sine and cosine transforms : principles and applications." Diss., Georgia Institute of Technology, 1993. http://hdl.handle.net/1853/15038.
Full text
9
Dušanka, Perišić. "On Integral Transforms and Convolution Equations on the Spaces of Tempered Ultradistributions." Phd thesis, Univerzitet u Novom Sadu, Prirodno-matematički fakultet u Novom Sadu, 1992. https://www.cris.uns.ac.rs/record.jsf?recordId=73337&source=NDLTD&language=en.
Full textAbstract:
In the thesis are introduced and investigated spaces of Burling and of Roumieu type tempered ultradistributions, which are natural generalization of the space of Schwartz’s tempered distributions in Denjoy-Carleman-Komatsu’s theory of ultradistributions. It has been proved that the introduced spaces preserve all of the good properties Schwartz space has, among others, a remarkable one, that the Fourier transform maps continuposly the spaces into themselves.In the first chapter the necessary notation and notions are given.In the second chapter, the spaces of ultrarapidly decreasing ultradifferentiable functions and their duals, the spaces of Beurling and of Roumieu tempered ultradistributions, are introduced; their topological properties and relations with the known distribution and ultradistribution spaces and structural properties are investigated; characterization of the Hermite expansions and boundary value representation of the elements of the spaces are given.The spaces of multipliers of the spaces of Beurling and of Roumieu type tempered ultradistributions are determined explicitly in the third chapter.The fourth chapter is devoted to the investigation of Fourier, Wigner, Bargmann and Hilbert transforms on the spaces of Beurling and of Roumieu type tempered ultradistributions and their test spaces.In the fifth chapter the equivalence of classical definitions of the convolution of Beurling type ultradistributions is proved, and the equivalence of, newly introduced definitions, of ultratempered convolutions of Beurling type ultradistributions is proved.In the last chapter is given a necessary and sufficient condition for a convolutor of a space of tempered ultradistributions to be hypoelliptic in a space of integrable ultradistribution, is given, and hypoelliptic convolution equations are studied in the spaces.Bibliograpy has 70 items.U ovoj tezi su proučavani prostori temperiranih ultradistribucija Beurlingovog i Roumieovog tipa, koji su prirodna uopštenja prostora Schwarzovih temperiranih distribucija u Denjoy-Carleman-Komatsuovoj teoriji ultradistribucija. Dokazano je ovi prostori imaju sva dobra svojstva, koja ima i Schwarzov prostor, izmedju ostalog, značajno svojstvo da Furijeova transformacija preslikava te prostore neprekidno na same sebe.U prvom poglavlju su uvedene neophodne oznake i pojmovi.U drugom poglavlju su uvedeni prostori ultrabrzo opadajucih ultradiferencijabilnih funkcija i njihovi duali, prostori Beurlingovih i Rumieuovih temperiranih ultradistribucija; proučavana su njihova topološka svojstva i veze sa poznatim prostorima distribucija i ultradistribucija, kao i strukturne osobine; date su i karakterizacije Ermitskih ekspanzija i graničnih reprezentacija elemenata tih prostora.Prostori multiplikatora Beurlingovih i Roumieuovih temperiranih ultradistribucija su okarakterisani u trećem poglavlju.Četvrto poglavlje je posvećeno proučavanju Fourierove, Wignerove, Bargmanove i Hilbertove transformacije na prostorima Beurlingovih i Rouimieovih temperiranih ultradistribucija i njihovim test prostorima.U petoj glavi je dokazana ekvivalentnost klasičnih definicija konvolucije na Beurlingovim prostorima ultradistribucija, kao i ekvivalentnost novouvedenih definicija ultratemperirane konvolucije ultradistribucija Beurlingovog tipa.U poslednjoj glavi je dat potreban i dovoljan uslov da konvolutor prostora temperiranih ultradistribucija bude hipoeliptičan u prostoru integrabilnih ultradistribucija i razmatrane su neke konvolucione jednačine u tom prostoru.Bibliografija ima 70 bibliografskih jedinica.
10
Marir, F. "The application of number theoretic transforms to two dimensional convolutions and adaptive filtering." Thesis, University of Newcastle Upon Tyne, 1986. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.374136.
Full text
11
Homeili, Saeid. "Metrological characterisation of Low Power Voltage Transformers by using impulse response analysis." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/20998/.
Full textAbstract:
this thesis presents a new approach in dealing with characterize LPVT and proposes determining the impulse response of LPVT, purposing to find transfer function (h(t)) which contains most electrical characteristics of LPVTs as a dynamic system.
12
Giulieri, Laurent. "Séparation aveugle de sources basée sur l'utilisation des transformées spatiales quadratiques." Toulon, 2003. http://www.theses.fr/2003TOUL0016.
Full textAbstract:
Le problème auquel on s'intéresse est l'étude de la séparation aveugle de sources dans le cadre des mélanges instantanés en utilisant des transformées spatiales quadratiques. Après une présentation de la problématique centrale et des principaux travaux s'y rapportant, les outils (diagonalisation et zéro-diagonalisation conjointe d'un ensemble de matrices, transformées bilinéaires et quadratiques avec comme cas particulier les représentationstemps-fréquence, extensions au cas vectoriel avec les transformées spatiales quadratiques) autour desquels s'articuleront les méthodes développées dans cette thèse, sont détaillés. Le travail original présenté dans ce manuscrit est l'étude des transformées spatiales quadratiques des mélanges de sources non bruités qui conduit à l'établissement de deux critères de sélection automatique des points temps-fréquence permettant la construction d'ensemble de matrices devant être diagonalisées et/ou zéro-diagonalisées conjointement. Le premier critère est destiné au cas des mélanges de signaux déterministes décorrélés préalablement blanchis, mélanges qui pourront être complexes. Le second critère, plus large, est destiné au cas des mélanges de signaux déterministes corrélés. L'étape de blanchiment est supprimée et les techniques de diagonalisation conjointe sous contrainte d'unitarité sont remplacées, au profit d'une technique sans contrainte, limitant les performances de séparation. On démontre qu'il est aussi pos-sible de résoudre le problème de manière analytique. Enfin l'effet du bruit sur les représentations temps-fréquence spatiales quadratiques est étudié afin d'élaborer de nouvelles stratégies de décision nécessaires à l'adaptation des critères de sélection automatique au cas bruitéWe are concerned with the blind source separation for instantaneous mixing problem using quadratic spatial trans-forms. The central problematics and these main works being presented, the tools (joint diagonalization and zero-diagonalization of a set of matrices, quadratic and bilinear transforms with the particular case of the time-frequency representations, extension to the vectorial case with the quadratic spatial transforms) used to developed methods are detailed. The study of quadratic spatial transform of mixed sources, in the noiseless case, represents the original work of this document. It leads to the design of the two automatic selection criteria of time-frequency points allowing to the elaboration of the set of matrices having to be joint diagonalized and/or joint zero-diagonalized. The first criterion is intended for the uncorrelated deterministic sources mixing, previously whitened, where the mixing could be complex. The second criterion, more general, is intended for the correlated deterministic sources mixing. The whitening step is removed and the joint diagonalization techniques are replaced by a technic without constraint, limiting the separation performances. We show that it is possible to solve the problem by an analytical way. Finally, the noise effect on quadratic spatial time-frequency representation is studied in order to design new decision strategies for the adaptation of the automatic selection criteria to the noisy case
13
Proença, Maria da Conceiçao Machado Sangreman. "La transformée en ondelettes appliquée au traitement d'images de télédétection." Toulouse 3, 1992. http://www.theses.fr/1992TOU30260.
Full textAbstract:
La diversite des capteurs utilisables pour l'etude de la biosphere continentale par satellite et l'emboitement des structures et des mecanismes de fonctionnement conduisent a une grande multiplicite d'echelles interessantes pour la perception des ecosystemes observes. L'utilisation conjointe d'images de teledetection ayant des resolutions spatiales differentes apparait alors indispensable. Dans ce contexte, l'objectif de la these etait de definir et d'evaluer les conditions dans lesquelles la transformee en ondelettes pouvait apporter des reponses methodologiques a l'analyse multiresolution. Le travail a consiste, dans une premiere phase, a effectuer une revue des principes mathematiques qui conduit aux choix de privilegier le traitement de donnees dans le domaine spatial. Le deuxieme chapitre presente ensuite les deux methodes implementees en y associant une evaluation quantitative des resultats a une et deux dimensions fournis par les algorithmes developpes. La troisieme partie decrit les resultats obtenus dans l'espace des approximations par la mise en uvre d'une methode de fusion de donnees utilisant la transformee en ondelettes discretes. Le dernier chapitre montre comment l'application de l'analyse en ondelettes continues permet d'ouvrir des possibilites nouvelles en analyse d'images: filtrage associe a l'extraction d'attributs et ameliorant des resultats de classification, segmentation sur des attributs de forme et de taille
14
Pavel, Dimovski. "Translation invariant Banach spaces of distributions and boundary values of integral transform." Phd thesis, Univerzitet u Novom Sadu, Prirodno-matematički fakultet u Novom Sadu, 2015. https://www.cris.uns.ac.rs/record.jsf?recordId=93767&source=NDLTD&language=en.
Full textAbstract:
We use common notation ∗ for distribution (Scshwartz), (Mp) (Beurling) i {Mp} (Roumieu) setting. We introduce and study new (ultra) distribution spaces, the test function spaces D∗E and their strong duals D'∗E’*.These spaces generalize the spaces D∗Lq , D'∗Lp , B’* and their weighted versions. The construction of our new (ultra)distribution spaces is based on the analysis of a suitable translation-invariant Banach space of (ultra)distribution E with continuous translation group, which turns out to be a convolution module over the Beurling algebra L1ω, where the weight ω is related to the translation operators on E. The Banach space E’∗ stands for L1ωˇ ∗ E’. We apply our results to the study of the convolution of ultradistributions. The spaces of convolutors O’∗C (Rn) for tempered ultradistributions are analyzed via the duality with respect to the test functionspaces O∗C (Rn), introduced in this thesis. Using the properties of translationinvariantBanach space of ultradistributions E we obtain a full characterization ofthe general convolution of Roumieu ultradistributions via the space of integrableultradistributions is obtained. We show: The convolution of two Roumieu ultradistributions T, S ∈ D’{Mp} (Rn) exists if and only if (φ ∗ Š) T ∈ D’{Mp}L1(Rn) for every φ ∈ D {Mp} (Rn). We study boundary values of holomorphic functions defined in tube domains. New edge of the wedge theorems are obtained. The resultsare then applied to represent D’E’* as a quotient space of holomorphic functions.We also give representations of elements of D’E’* via the heat kernel method.Koristimo oznaku ∗ za distribuciono (Svarcovo), (Mp) (Berlingovo) i {Mp} (Roumieuovo) okruženje. Uvodimo i prouavamo nove (ultra)distribucione prostore, test funkcijske prostore D∗E i njihove duale D'∗E'*. Ovi prostori uopštavaju prostore D∗Lq , D'∗Lp , B'∗ i njihove težinske verzije. Konstrukcija naših novih (ultra)distribucionih prostora je zasnovana na analizi odgovarajuićh translaciono - invarijantnih Banahovih prostora (ultra)distribucija koje označavamo sa E. Ovi prostori imaju neprekidnu grupu translacija, koja je konvolucioni modul nad Beurlingovom algebrom L1ω, gde je težina ω povezana sa operatorima translacije prostora E. Banahov prostor E'∗ označava prostor L1ω˅ ∗ E'. Koristeći dobijene rezultata proučavamo konvoluciju ultradistribucija. Prostori konvolutora O'∗C (Rn) temperiranih ultradistribucija, analizirani su pomoću dualnosti test funkcijskih prostora O∗C (Rn), definisanih u ovoj tezi. Koristeći svojstva translaciono - invarijantnih Banahovih prostora temperiranih ultradistribucija, opet označenih sa E, dobijamo karakterizaciju konvolucije Romuieu-ovih ultradistribucija, preko integrabilnih ultradistribucija. Dokazujemo da: konvolucija dve Roumieu-ove ultradistribucija T, S ∈ D'{Mp} (Rn) postoji ako i samo ako (φ ∗ Sˇ)T ∈ D'{Mp} L1 (Rn) za svaki φ ∈ D{Mp}(Rn). Takođe, proučavamo granične vrednosti holomorfnih funkcija definisanih na tubama. Dokazane su nove teoreme ”otrog klina”. Rezultati se zatim koriste za prezentaciju D'E'∗ preko faktor prostora holomorfnih funkcija. Takođe, data je prezentacija elemente D'E'∗ koristeći heat kernel metode.
15
Sazish, Abdul Naser. "Efficient architectures and power modelling of multiresolution analysis algorithms on FPGA." Thesis, Brunel University, 2011. http://bura.brunel.ac.uk/handle/2438/6290.
Full textAbstract:
In the past two decades, there has been huge amount of interest in Multiresolution Analysis Algorithms (MAAs) and their applications. Processing some of their applications such as medical imaging are computationally intensive, power hungry and requires large amount of memory which cause a high demand for efficient algorithm implementation, low power architecture and acceleration. Recently, some MAAs such as Finite Ridgelet Transform (FRIT) Haar Wavelet Transform (HWT) are became very popular and they are suitable for a number of image processing applications such as detection of line singularities and contiguous edges, edge detection (useful for compression and feature detection), medical image denoising and segmentation. Efficient hardware implementation and acceleration of these algorithms particularly when addressing large problems are becoming very chal-lenging and consume lot of power which leads to a number of issues including mobility, reliability concerns. To overcome the computation problems, Field Programmable Gate Arrays (FPGAs) are the technology of choice for accelerating computationally intensive applications due to their high performance. Addressing the power issue requires optimi- sation and awareness at all level of abstractions in the design flow. The most important achievements of the work presented in this thesis are summarised here. Two factorisation methodologies for HWT which are called HWT Factorisation Method1 and (HWTFM1) and HWT Factorasation Method2 (HWTFM2) have been explored to increase number of zeros and reduce hardware resources. In addition, two novel efficient and optimised architectures for proposed methodologies based on Distributed Arithmetic (DA) principles have been proposed. The evaluation of the architectural results have shown that the proposed architectures results have reduced the arithmetics calculation (additions/subtractions) by 33% and 25% respectively compared to direct implementa-tion of HWT and outperformed existing results in place. The proposed HWTFM2 is implemented on advanced and low power FPGA devices using Handel-C language. The FPGAs implementation results have outperformed other existing results in terms of area and maximum frequency. In addition, a novel efficient architecture for Finite Radon Trans-form (FRAT) has also been proposed. The proposed architecture is integrated with the developed HWT architecture to build an optimised architecture for FRIT. Strategies such as parallelism and pipelining have been deployed at the architectural level for efficient im-plementation on different FPGA devices. The proposed FRIT architecture performance has been evaluated and the results outperformed some other existing architecture in place. Both FRAT and FRIT architectures have been implemented on FPGAs using Handel-C language. The evaluation of both architectures have shown that the obtained results out-performed existing results in place by almost 10% in terms of frequency and area. The proposed architectures are also applied on image data (256 £ 256) and their Peak Signal to Noise Ratio (PSNR) is evaluated for quality purposes. Two architectures for cyclic convolution based on systolic array using parallelism and pipelining which can be used as the main building block for the proposed FRIT architec-ture have been proposed. The first proposed architecture is a linear systolic array with pipelining process and the second architecture is a systolic array with parallel process. The second architecture reduces the number of registers by 42% compare to first architec-ture and both architectures outperformed other existing results in place. The proposed pipelined architecture has been implemented on different FPGA devices with vector size (N) 4,8,16,32 and word-length (W=8). The implementation results have shown a signifi-cant improvement and outperformed other existing results in place. Ultimately, an in-depth evaluation of a high level power macromodelling technique for design space exploration and characterisation of custom IP cores for FPGAs, called func-tional level power modelling approach have been presented. The mathematical techniques that form the basis of the proposed power modeling has been validated by a range of custom IP cores. The proposed power modelling is scalable, platform independent and compares favorably with existing approaches. A hybrid, top-down design flow paradigm integrating functional level power modelling with commercially available design tools for systematic optimisation of IP cores has also been developed. The in-depth evaluation of this tool enables us to observe the behavior of different custom IP cores in terms of power consumption and accuracy using different design methodologies and arithmetic techniques on virous FPGA platforms. Based on the results achieved, the proposed model accuracy is almost 99% true for all IP core's Dynamic Power (DP) components.
16
Gabriel, Eric [Verfasser]. "Automatic Multi-Scale and Multi-Object Pedestrian and Car Detection in Digital Images Based on the Discriminative Generalized Hough Transform and Deep Convolutional Neural Networks / Eric Gabriel." Kiel : Universitätsbibliothek Kiel, 2019. http://d-nb.info/1187732826/34.
Full text
17
Hammarqvist, Ulf. "Audio editing in the time-frequency domain using the Gabor Wavelet Transform." Thesis, Uppsala universitet, Centrum för bildanalys, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-153634.
Full textAbstract:
Visualization, processing and editing of audio, directly on a time-frequency surface, is the scope of this thesis. More precisely the scalogram produced by a Gabor Wavelet transform is used, which is a powerful alternative to traditional techinques where the wave form is the main visual aid and editting is performed by parametric filters. Reconstruction properties, scalogram design and enhancements as well audio manipulation algorithms are investigated for this audio representation.The scalogram is designed to allow a flexible choice of time-frequency ratio, while maintaining high quality reconstruction. For this mean, the Loglet is used, which is observed to be the most suitable filter choice. Re-assignmentare tested, and a novel weighting function using partial derivatives of phase is proposed. An audio interpolation procedure is developed and shown to perform well in listening tests.The feasibility to use the transform coefficients directly for various purposes is investigated. It is concluded that Pitch shifts are hard to describe in the framework while noise thresh holding works well. A downsampling scheme is suggested that saves on operations and memory consumption as well as it speeds up real world implementations significantly. Finally, a Scalogram 'compression' procedure is developed, allowing the caching of an approximate scalogram.
18
Le, Van Linh. "Automatic landmarking for 2D biological images : image processing with and without deep learning methods." Thesis, Bordeaux, 2019. http://www.theses.fr/2019BORD0238.
Full textAbstract:
Les points de repère sont présentés dans les applications de différents domaines tels que le biomédical ou le biologique. C’est également l’un des types de données qui ont été utilisés dans différentes analyses, par exemple, ils ne sont pas seulement utilisés pour mesurer la forme de l’objet, mais également pour déterminer la similarité entre deux objets. En biologie, les repères sont utilisés pour analyser les variations inter-organismes. Cependant, l’offre de repères est très lourde et le plus souvent, ils sont fournis manuellement. Ces dernières années, plusieurs méthodes ont été proposées pour prédire automatiquement les points de repère, mais la dureté existe, car elles se sont concentrées sur des données spécifiques. Cette thèse porte sur la détermination automatique de points de repère sur des images biologiques, plus spécifiquement sur d’images 2D des coléoptères. Dans le cadre de nos recherches, nous avons collaboré avec des biologistes pour créer un ensemble de données comprenantles images de 293 coléoptères. Pour chaque coléoptère de cette donnée, 5 images corre-spondent à 5 parties prises en compte, par exemple tête, élytre, pronotum, mandibule gauche et droite. Avec chaque image, un ensemble de points de repère a été proposé manuellement par les biologistes. La première étape, nous avons apporté une méthode qui a été appliquée sur les ailes de mouche, à appliquer sur notre jeu de données dans le but de tester la pertinence des techniques de traitement d’image sur notre problème. Deuxièmement, nous avons développé une méthode en plusieurs étapes pour fournir automatiquement les points de repère sur les images. Ces deux premières étapes ont été effectuées sur les images de la mandibule qui sont considérées comme évidentes pour l’utilisation des méthodes de traitement d’images. Troisièmement, nous avons continué à considérer d’autres parties complexes restantes de coléoptères. En conséquence, nous avons utilisé l’aide de Deep Learning. Nous avons conçu un nouveau modèle de Convolutional Neural Network, nommé EB-Net, pour prédire les points de repère sur les images restantes. De plus, nous avons proposé une nouvelle procédurepour augmenter le nombre d’images dans notre jeu de données, ce qui est considéré comme notre limite à appliquer Deep Learning. Enfin, pour améliorer la qualité des coordonnées prédites, nous avons utilisé Transfer Learning, une autre technique de Deep Learning. Pour ce faire, nous avons formé EB-Net sur les points clés publics du visage. Ensuite, ils ont été transférés pour affiner les images de coléoptère. Les résultats obtenus ont été discutés avec les biologistes et ils ont confirmé que la qualité des repéres prédits est suffisamment bonne sur la plane statistique pour remplacer les repères manuels pour la plupart des analyses de morphométrie différentesLandmarks are presented in the applications of different domains such as biomedical or biological. It is also one of the data types which have been usedin different analysis, for example, they are not only used for measuring the form of the object, but also for determining the similarity between two objects. In biology, landmarks are used to analyze the inter-organisms variations, however the supply of landmarks is very heavy and most often they are provided manually. In recent years, several methods have been proposed to automatically predict landmarks, but it is existing the hardness because these methods focused on the specific data. This thesis focuses on automatic determination of landmarks on biological images, more specifically on two-dimensional images of beetles. In our research, we have collaborated with biologists to build a dataset including the images of 293 beetles. For each beetle in this dataset, 5 images correspond to 5 parts have been taken into account, e.g., head, body, pronotum, left and right mandible. Along with each image, a set of landmarks has been manually proposed by biologists. First step, we have brought a method whichwas applied on fly wings, to apply on our dataset with the aim to test the suitability of image processing techniques on our problem. Secondly, we have developed a method consisting of several stages to automatically provide the landmarks on the images.These two first steps have been done on the mandible images which are considered as obvious to use the image processing methods. Thirdly, we have continued to consider other complex remaining parts of beetles. Accordingly, we have used the help of Deep Learning. We have designed a new model of Convolutional Neural Network, named EB-Net, to predict the landmarks on remaining images. In addition, we have proposed a new procedure to augment the number of images in our dataset, which is seen as our limitation to apply deep learning. Finally, to improve the quality of predicted coordinates, we have employed Transfer Learning, another technique of Deep Learning. In order to do that, we trained EB-Net on a public facial keypoints. Then, they were transferred to fine-tuning on beetle’s images. The obtained results have been discussed with biologists, and they have confirmed that the quality of predicted landmarks is statistically good enough to replace the manual landmarks for most of the different morphometry analysis
19
Bertani, Federico. "Deep Learning methods for Portfolio Optimization." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/24245/.
Full textAbstract:
Portfolio optimization is one of the most studied fields that have been researched with machine learning approaches because of its inherent demand for forecasting future market properties. In this thesis, it is shown how one can use deep neural networks with historical returns to do risk adjusted asset allocation. Unlike previous studies which set as target variable asset prices, the variable to predict here is represented by the best asset allocation strategy. Experiments performed on a time period of seven years show that temporal convolutional networks are superior to long short term memory networks and transformers. Compared to baseline benchmarks, the computed allocation has an average increase in the year revenue between 2% and 5%. Furthermore, results are compared against equally weighted, inverse volatility and risk parity methods, showing higher cumulative returns than the first method and equal if not higher in some cases than the latter methods.
20
Jatobá, Ariosvaldo Marques. "Tipos de holomorfia em espaços de Banach." [s.n.], 2008. http://repositorio.unicamp.br/jspui/handle/REPOSIP/307327.
Full textAbstract:
Orientador: Jorge Tulio Mujica AscuiTese (doutorado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica
Made available in DSpace on 2018-08-12T04:40:55Z (GMT). No. of bitstreams: 1 Jatoba_AriosvaldoMarques_D.pdf: 617454 bytes, checksum: b1f5fca80565a5a0f7b273ed5238b6ee (MD5) Previous issue date: 2008
Resumo: Neste trabalho introduzimos e estudamos os espaços das funções inteiras ?-holomorfas de tipo limitado. Em particular obtemos resultados de dualidade via a transformada de Borel e provamos resultados de existência e aproximação para equações de convolução. Os resultados provados generalizam resultados anteriores deste tipo devido a C. Gupta [21], M. Matos [28] e X. Mujica [32]. Nós estudamos as relações entre o espaço Hb(E; F) das funções inteiras de tipo limitado, o espaço HNb(E; F) das funções inteiras nucleares de tipo limitado, o espaço HPIb(E; F) das funções inteiras Pietsch-integrais de tipo limitado, e o espaço HGIb (E; F) das funções inteiras Grothendieck-integrais de tipo limitado. Estendemos para o caso de funções inteiras resultados de R. Alencar [2] e R. Cilia e J. Gutierrez [10] no caso de polinômios homogêneos.
Abstract: In this work we introduce and study functions of ?-holomorphy type of bounded type. In particular we obtain a duality result via the Borel transform and we prove existence and approximation results for convolution equations. The results we prove generalize previous results of this type due to C. Gupta [21], M. Matos [28] and X. Mujica [32]. We study the relationships among the space Hb(E; F) of entire mappings of bounded type, the space HNb(E; F) of entire mappings of nuclear bounded type, the space HPIb(E; F) of entire mappings of Pietsch integral bounded type, and the space HGIb(E; F) of entire mappings of Grothendieck integral bounded type. We extend to the case of entire mappings several results due to R. Alencar [2] and R. Cilia and J. Gutiérrez [10] in the case of homogeneous polynomials.
Doutorado
Analise Funcional
Doutor em Matemática
21
Kloeckner, Jonas. "Tabela de covariância : um mapeamento rápido e automático de continuidade espacial." reponame:Biblioteca Digital de Teses e Dissertações da UFRGS, 2018. http://hdl.handle.net/10183/184705.
Full textAbstract:
Os modelos de covariância são ferramentas geoestatísticas essenciais para mapear a continuidade espacial. A abordagem atual busca um modelo de continuidade espacial lícito com mínima ou até mesmo sem nenhuma interferência do usuário. Alinhado a essa visão moderna, é proposto obter uma tabela de covariância que visa substituir na prática o modelo tradicional explicitamente definido de covariância. Essa tabela de covariância é obtida por meio de três etapas: interpolar o conjunto de dados para preencher um grid regular, aplicar a convolução através do algoritmo da transformada rápida de Fourier e, por fim, transformar de volta para o domínio espacial. O modelo base para extrair covariância representa o ponto chave comparando com os métodos anteriores que propuseram o uso da tabela de covariância. Os resultados são satisfatórios, tanto na validação estatística do método, quanto na rapidez de obtenção de uma análise de continuidade espacial. Um estudo de caso tridimensional ilustra a aplicação prática através de krigagem e simulação geoestatística em comparação com a modelagem espacial tradicional.Covariance models are essential geostatistical tools to map spatial continuity. The current approach pursues a licit spatial continuity model with minimum or even no user interference. Aligned with this modern view we propose to obtain a covariance table that aims at replacing in practice traditional covariance explicit defined model. This covariance table is obtained through a three steps work flow: interpolating the dataset to fill up a regular grid, auto convolute via Fast Fourier Transform algorithm and back transform to spacial domain. The base model to extract covariance represents the turning point comparing with previous methods that proposed covariance table usage. The results are satisfactory, both in the statistical validation of the method and in the speed of obtaining a spatial continuity analysis. A three dimensional case study illustrates the practical application for kriging and geostatistical simulation in comparison with traditional spatial modeling.
22
Nassar, Alaa S. N. "A Hybrid Multibiometric System for Personal Identification Based on Face and Iris Traits. The Development of an automated computer system for the identification of humans by integrating facial and iris features using Localization, Feature Extraction, Handcrafted and Deep learning Techniques." Thesis, University of Bradford, 2018. http://hdl.handle.net/10454/16917.
Full textAbstract:
Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level.
Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image.
Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity.Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. This PhD thesis is focused on the combination of both the face and the left and right irises, in a unified hybrid multimodal biometric identification system using different fusion approaches at the score and rank level.
Firstly, the facial features are extracted using a novel multimodal local feature extraction approach, termed as the Curvelet-Fractal approach, which based on merging the advantages of the Curvelet transform with Fractal dimension. Secondly, a novel framework based on merging the advantages of the local handcrafted feature descriptors with the deep learning approaches is proposed, Multimodal Deep Face Recognition (MDFR) framework, to address the face recognition problem in unconstrained conditions. Thirdly, an efficient deep learning system is employed, termed as IrisConvNet, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image.
Finally, The performance of the unimodal and multimodal systems has been evaluated by conducting a number of extensive experiments on large-scale unimodal databases: FERET, CAS-PEAL-R1, LFW, CASIA-Iris-V1, CASIA-Iris-V3 Interval, MMU1 and IITD and MMU1, and SDUMLA-HMT multimodal dataset. The results obtained have demonstrated the superiority of the proposed systems compared to the previous works by achieving new state-of-the-art recognition rates on all the employed datasets with less time required to recognize the person’s identity.Higher Committee for Education Development in Iraq
23
Puigt, Matthieu. "Méthodes de séparation aveugle de sources fondées sur des transformées temps-fréquence : application à des signaux de parole." Toulouse 3, 2007. http://thesesups.ups-tlse.fr/217/.
Full textAbstract:
Plusieurs méthodes de séparation aveugle de source (SAS), fondées sur des transformées temps-fréquence (TF), ont été proposées au cours de cette thèse. En sortie des systèmes utilisés, une contribution de chaque source est estimée, uniquement à l'aide des signaux mélangés. Toutes les méthodes étudiées dans ce manuscrit trouvent des petites zones du plan TF où une seule source est présente et estiment dans ces zones les paramètres de mélange. Ces approches sont particulièrement adaptées aux sources non-stationnaires. Nous avons tout d'abord étudié et amélioré des méthodes proposées précédemment par l'équipe, basées sur des critères de variance ou de corrélation, pour des mélanges linéaires instantanés. Elles apportent d'excellentes performances pour des signaux de parole et peuvent aussi séparer des spectres issus de données astrophysiques. Cependant, la nature des mélanges qu'elles peuvent traiter limite leur champ d'application. Nous avons donc étendu ces approches à des mélanges plus réalistes. Les premières extensions considèrent des mélanges de sources atténuées et décalées temporellement, ce qui correspond physiquement aux mélanges en chambre anéchoïque. Elles nécessitent des hypothèses de parcimonie beaucoup moins fortes que certaines approches de la littérature, tout en traitant le même type de mélanges. Nous avons étudié l'apport de méthodes de classification non-supervisée sur nos approches et avons obtenu de bonnes performances pour des mélanges de signaux de parole. Enfin, une extension théorique aux mélanges convolutifs généraux est décrite mais nécessite de fortes hypothèses de parcimonie et le réglage d'indéterminations propres aux méthodes fréquentiellesSeveral time-frequency (TF) blind source separation (BSS) methods have been proposed in this thesis. In the systems output that have been used, a contribution of each source is estimated, using only mixed signals. All the methods proposed in this manuscript find tiny TF zones where only one source is active and estimate the mixing parameters in these zones. These approaches are particularly well suited for non-stationary sources (speech, music). We first studied and improved linear instantaneous methods based on variance or correlation criteria, that have been previously proposed by our team. They yield excellent performance for signals of speech and can also separate spectra from astrophysical data. However, the nature of the mixtures that they can process limits their application fields. We have extended these approaches to more realistic mixtures. The first extensions consider attenuated and delayed mixtures of sources, which corresponds to mixtures in anechoic chamber. They require less restrictive sparsity assumptions than some approaches previously proposed in the literature, while addressing the same type of mixtures. We have studied the contribution of clustering techniques to our approaches and have achieved good performance for mixtures of speech signals. Lastly, a theoretical extension of these methods to general convolutive mixtures is described. It needs strong sparsity hypotheses and we have to solve classical indeterminacies of frequency-domain BSS methods
24
Herrmann, Felix J. "Surface related multiple prediction from incomplete data." European Association of Geoscientists & Engineers, 2007. http://hdl.handle.net/2429/551.
Full textAbstract:
Incomplete data, unknown source-receiver signatures and free-surface reflectivity represent
challenges for a successful prediction and subsequent removal of multiples. In
this paper, a new method will be represented that tackles these challenges by combining
what we know about wavefield (de-)focussing, by weighted convolutions/correlations,
and recently developed curvelet-based recovery by sparsity-promoting inversion (CRSI).
With this combination, we are able to leverage recent insights from wave physics towards
a nonlinear formulation for the multiple-prediction problem that works for incomplete
data and without detailed knowledge on the surface effects.
25
Rajbhandari, Samyam. "Locality Optimizations for Regular and Irregular Applications." The Ohio State University, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=osu1469033289.
Full text
26
Benaych-Georges, Florent. "Matrices aléatoires et probabilités libres." Paris 6, 2005. http://www.theses.fr/2005PA066566.
Full text
27
Dall'Olio, Lorenzo. "Estimation of biological vascular ageing via photoplethysmography: a comparison between statistical learning and deep learning." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/21687/.
Full textAbstract:
This work aims to exploit the biological ageing phenomena which affects human blood vessels. The analysis is performed starting from a database of photoplethysmographic signals acquired through smartphones. The further step involves a preprocessing phase, where the signals are detrended using a central moving average filter, demoduled using the envelope of the analytic signal obtained from the Hilbert transform, denoised using the central moving average filter over the envelope. After the preprocessing we compared two different approaches. The first one regards Statistical Learning, which involves feature extraction and selection through the usage of statistics and machine learning algorithms. This in order to perform a classification supervised task over the chronological age of the individual, which is used as a proxy for healthy/non healthy vascular ageing. The second one regards Deep Learning, which involves the realisation of a convolutional neural network to perform the same task, but avoiding the feature extraction/selection step and so possible bias introduced by such phases. Doing so we obtained comparable outcomes in terms of area under the curve metrics from a 12 layers ResNet convolutional network and a support vector machine using just covariates together with a couple of extracted features, acquiring clues regarding the possible usage of such features as biomarkers for the vascular ageing process. The two mentioned features can be related with increasing arterial stiffness and increasing signal randomness due to ageing.
28
Moravec, Zbyněk. "Kontrola zobrazení textu ve formulářích." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2017. http://www.nusl.cz/ntk/nusl-363844.
Full textAbstract:
Purpose of this thesis is the quality check of correct button text display on photographed monitors. These photographs contain a variety of image distortions which complicates the following image graphic element recognition. This paper outlines several possibilities to detect buttons on forms and further elaborates on the implemented detection based on contour shapes description. After buttons are found, their defects are detected subsequently. Additionally, this thesis describes an automatic identification of picture with the highest quality for documentation purposes.
29
Kaalen, Stefan. "Semi-Markov processes for calculating the safety of autonomous vehicles." Thesis, KTH, Matematisk statistik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-252331.
Full textAbstract:
Several manufacturers of road vehicles today are working on developing autonomous vehicles. One subject that is often up for discussion when it comes to integrating autonomous road vehicles into the infrastructure is the safety aspect. There is in the context no common view of how safety should be quantified. As a contribution to this discussion we propose describing each potential hazardous event of a vehicle as a Semi-Markov Process (SMP). A reliability-based method for using the semi-Markov representation to calculate the probability of a hazardous event to occur is presented. The method simplifies the expression for the reliability using the Laplace-Stieltjes transform and calculates the transform of the reliability exactly. Numerical inversion algorithms are then applied to approximate the reliability up to a desired error tolerance. The method is validated using alternative techniques and is thereafter applied to a system for automated steering based on a real example from the industry. A desired evolution of the method is to involve a framework for how to represent each hazardous event as a SMP.Flertalet tillverkare av vägfordon jobbar idag på att utveckla autonoma fordon. Ett ämne ofta på agendan i diskussionen om att integrera autonoma fordon på vägarna är säkerhet. Det finns i sammanhanget ingen klar bild över hur säkerhet ska kvantifieras. Som ett bidrag till denna diskussion föreslås här att beskriva varje potentiellt farlig situation av ett fordon som en Semi-Markov process (SMP). En metod presenteras för att via beräkning av funktionssäkerheten nyttja semi-Markov representationen för att beräkna sannolikheten för att en farlig situation ska uppstå. Metoden nyttjar Laplace-Stieltjes transformen för att förenkla uttrycket för funktionssäkerheten och beräknar transformen av funktionssäkerheten exakt. Numeriska algoritmer för den inversa transformen appliceras sedan för att beräkna funktionssäkerheten upp till en viss feltolerans. Metoden valideras genom alternativa tekniker och appliceras sedan på ett system för autonom styrning baserat på ett riktigt exempel från industrin. En fördelaktig utveckling av metoden som presenteras här skulle vara att involvera ett ramverk för hur varje potentiellt farlig situation ska representeras som en SMP.
30
Limouzineau, Maÿlis. "Constructions de sous-variétés legendriennes dans les espaces de jets d'ordre un de fonctions et fonctions génératrices." Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066676/document.
Full textAbstract:
Dans cette thèse, on manipule deux types d'objets fondamentaux de la topologie de contact : les sous-variétés legendriennes des espaces de 1-jets de fonctions dé finies sur une variété M, noté J1(M;R), et la notion intimement liée de fonctions génératrices. On étudie des "opérations" que l'on peut faire sur ces objets, c'est-à-dire des procédures qui construisent (génériquement) de nouvelles sous-variétés legendriennes à partir d'anciennes. On dé finit en particulier les opérations somme et convolution des sous-variétés legendriennes, qui sont conjuguées par une transformation de type transformée de Legendre. Nous montrons que ces opérations se refl ètent harmonieusement dans le monde des fonctions génératrices. Ce second point de vue nous conduit en particulier à nous interroger sur l'effet de nos opérations sur le sélecteur, notion classique de géométrie symplectique dont on adapte la construction à ce contexte. Pour fi nir, on se concentre sur l'espace à trois dimensions J1(R;R) et sur les noeuds legendriens qui admettent (globalement) une fonction génératrice. C'est une condition forte sur les sous-variétés legendriennes, que l'on choisit d'étudier en proposant plusieurs constructions explicites. On termine avec l'étude des notions de cobordisme legendrien naturellement associées, où l'opération somme évoquée plus s'avère tenir une place centraleThis thesis concerns two types of fundamental objects of the contact topology : Legendrian submanifolds in 1-jet spaces of functions de fined on a manifold M, denoted by J1(M;R), and the closed related notion of generating functions. We study "operations" that build (generically) new Legendrian submanifolds from old ones. In particular, we de fined the operations sum and convolution of Legendrian submanifolds, which are linked by a form of the Legendre transform. We show how the operations are well re flected in terms of generating functions. It offers a second point of view and leads us to wonder the effect of our operations on the selector, which is a classical notion of symplectic geometry, and we adapt its construction to this context. Finally, we focus on the three dimensional space J1(R;R) and Legendrian knots which admit a (global) generating function. It is a strong condition for Legendrian submanifolds, and we choose to examine it by proposing several explicit constructions. We conclude by studying the notions of Legendrian cobordism which are naturally related. The operation sum mentioned before finds there a central role
31
Farias, Filho Antonio Pereira de. "A Transformada Discreta de Fourier no círculo finito ℤ/nℤ." Universidade Federal da Paraíba, 2016. http://tede.biblioteca.ufpb.br:8080/handle/tede/9429.
Full textAbstract:
Submitted by ANA KARLA PEREIRA RODRIGUES (anakarla_@hotmail.com) on 2017-09-05T12:56:54Z
No. of bitstreams: 1
arquivototal.pdf: 2044930 bytes, checksum: 05bad0799c40d5bf256cf504f0a8b5ab (MD5)Approved for entry into archive by ANA KARLA PEREIRA RODRIGUES (anakarla_@hotmail.com) on 2017-09-05T15:29:04Z (GMT) No. of bitstreams: 1 arquivototal.pdf: 2044930 bytes, checksum: 05bad0799c40d5bf256cf504f0a8b5ab (MD5)
Made available in DSpace on 2017-09-05T15:29:04Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 2044930 bytes, checksum: 05bad0799c40d5bf256cf504f0a8b5ab (MD5) Previous issue date: 2016-08-26
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES
We will do here a theoretical study of the Discrete Fourier Transform on the finite circle ℤ/nℤ. Our main objective is to see if we can get properties analogous to those found in the Fourier transform for the continuous case. In this work we show that ℤ/nℤ has a ring structure, providing conditions for the development of extensively discussed topics in arithmetic, for example, The Chinese Remainder Theorem, Euler’s Phi Function and primitive roots, themes these to be dealt with in first chapter. The main subject of this study is developed in the second chapter, which define the space L2(ℤ/nℤ) and prove that this is a finite-dimensional inner product vector space, with an orthonormal basis. This fact is of utmost importance when we are determining the matrix and demonstrating the properties of the discrete Fourier transform. We will also make geometric interpretations of the Chinese Remainder Theorem and the finite circle ℤ/nℤ as well as give a graphical representation of the DFT of some functions that calculate. During the development of this study we will make recurrent use of definitions and results treated in Arithmetic, Algebra and Linear Algebra.
Faremos, aqui, um estudo teórico sobre a Transformada Discreta de Fourier no círculo finito ℤ/nℤ. Nosso principal objetivo é verificar se podemos obter propriedades análogas às encontradas nas transformadas de Fourier para o caso contínuo. Nesse trabalho mostraremos que ℤ/nℤ tem uma estrutura de anel, dando condições para o desenvolvimento de temas bastante discutidos na Aritmética como, por exemplo, o Teorema Chinês do Resto, função Phi de Euler e raízes primitivas, temas estes que serão tratados no primeiro capítulo. O assunto principal desse estudo é desenvolvido no segundo capítulo, onde definiremos o espaço L2(ℤ/nℤ) e provaremos que este é um espaço vetorial com produto interno, dimensão finita e uma base ortonormal. Tal fato será de extrema importância quando estivermos determinando a matriz e demonstrando as propriedades da transformada discreta de Fourier. Também faremos interpretações geométricas do Teorema Chinês do Resto e do círculo finito ℤ/nℤ assim como daremos a representação gráfica da DFT de algumas funções que calcularemos. Durante o desenvolvimento desse estudo faremos uso recorrente de definições e resultados tratados na Aritmética, Álgebra e Álgebra Linear.
32
Talavašek, Kamil. "Srovnání implementací diskrétní waveletové transformace v Javě a C/C++." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2008. http://www.nusl.cz/ntk/nusl-217186.
Full textAbstract:
This thesis deals with discrete wavelet transform of a two dimensional discrete signal using the CDF9/7 wavelet. It contains theoretical principles of discrete wavelet transformation, a description of the convolution and Lifting calculation principle, characteristics of the CDF9/7 wavelet, the possibilities of processing a two dimensional input signal and boundary handling. Both principles (convolution, Lifting) of the calculation of two dimensional discrete wavelet transformation have been implemented within the thesis using the CDF9/7 wavelet in the C++ and Java programming languages. Four applications have been created that way which have been described in detail and then compared with one another especially in light of the calculation speed of discrete wavelet transformation. Finally an analysis of the acquired results and an evaluation of the whole thesis has been performed.
33
Papoutsellis, Evangelos. "First-order gradient regularisation methods for image restoration : reconstruction of tomographic images with thin structures and denoising piecewise affine images." Thesis, University of Cambridge, 2016. https://www.repository.cam.ac.uk/handle/1810/256216.
Full textAbstract:
The focus of this thesis is variational image restoration techniques that involve novel non-smooth first-order gradient regularisers: Total Variation (TV) regularisation in image and data space for reconstruction of thin structures from PET data and regularisers given by an infimal-convolution of TV and $L^p$ seminorms for denoising images with piecewise affine structures. In the first part of this thesis, we present a novel variational model for PET reconstruction. During a PET scan, we encounter two different spaces: the sinogram space that consists of all the PET data collected from the detectors and the image space where the reconstruction of the unknown density is finally obtained. Unlike most of the state of the art reconstruction methods in which an appropriate regulariser is designed in the image space only, we introduce a new variational method incorporating regularisation in image and sinogram space. In particular, the corresponding minimisation problem is formed by a total variational regularisation on both the sinogram and the image and with a suitable weighted $L^2$ fidelity term, which serves as an approximation to the Poisson noise model for PET. We establish the well-posedness of this new model for functions of Bounded Variation (BV) and perform an error analysis through the notion of the Bregman distance. We examine analytically how TV regularisation on the sinogram affects the reconstructed image especially the boundaries of objects in the image. This analysis motivates the use of a combined regularisation principally for reconstructing images with thin structures. In the second part of this thesis we propose a first-order regulariser that is a combination of the total variation and $L^p$ seminorms with $1 < p \le \infty$. A well-posedness analysis is presented and a detailed study of the one dimensional model is performed by computing exact solutions for simple functions such as the step function and a piecewise affine function, for the regulariser with $p = 2$ and $p = 1$. We derive necessary and sufficient conditions for a pair in $BV \times L^p$ to be a solution for our proposed model and determine the structure of solutions dependent on the value of $p$. In the case $p = 2$, we show that the regulariser is equivalent to the Huber-type variant of total variation regularisation. Moreover, there is a certain class of one dimensional data functions for which the regularised solutions are equivalent to high-order regularisers such as the state of the art total generalised variation (TGV) model. The key assets of our regulariser are the elimination of the staircasing effect - a well-known disadvantage of total variation regularisation - the capability of obtaining piecewise affine structures for $p = 1$ and qualitatively comparable results to TGV. In addition, our first-order $TVL^p$ regulariser is capable of preserving spike-like structures that TGV is forced to smooth. The numerical solution of the proposed first-order model is in general computationally more efficient compared to high-order approaches.
34
Erdogan, Ahmet Yasin. "Analysis of the effects of phase noise and frequency offset in orthogonal frequency division multiplexing (OFDM) systems." Thesis, Monterey California. Naval Postgraduate School, 2004. http://hdl.handle.net/10945/1712.
Full textAbstract:
Approved for public release, distribution is unlimitedOrthogonal frequency division multiplexing (OFDM) is being successfully used in numerous applications. It was chosen for IEEE 802.11a wireless local area network (WLAN) standard, and it is being considered for the fourthgeneration mobile communication systems. Along with its many attractive features, OFDM has some principal drawbacks. Sensitivity to frequency errors is the most dominant of these drawbacks. In this thesis, the frequency offset and phase noise effects on OFDM based communication systems are investigated under a variety of channel conditions covering both indoor and outdoor environments. The simulation performance results of the OFDM system for these channels are presented.
Lieutenant Junior Grade, Turkish Navy
35
Kopečný, Josef. "Návrh nové metody pro stereovidění." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2008. http://www.nusl.cz/ntk/nusl-235903.
Full textAbstract:
This thesis covers with the problems of photogrammetry. It describes the instruments, theoretical background and procedures of acquiring, preprocessing, segmentation of input images and of the depth map calculating. The main content of this thesis is the description of the new method of stereovision. Its algorithm, implementation and evaluation of experiments. The covered method belongs to correlation based methods. The main emphasis lies in the segmentation, which supports the depth map calculation.
36
Málková, Eliška. "Monitorování pohybu více objektů užitím křížové korelace." Master's thesis, Vysoké učení technické v Brně. Fakulta strojního inženýrství, 2019. http://www.nusl.cz/ntk/nusl-400456.
Full textAbstract:
Tato práce popisuje metodu analýzy translačního pohybu užitím křížové korelace. Ukazujeme, jakým způsobem se chová funkce křížové korelace obrazů s navzájem posunutými objekty, a jak nám to umožňuje nacházet jejich vektory posunu. Pro následnou implementaci je nalezena efektivní metoda pro hledání pouze požadovaného počtu lokálních maxim funkce.
37
Kornfeil, Vojtěch. "Soubor úloh pro kurs Sběr, analýza a zpracování dat." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2008. http://www.nusl.cz/ntk/nusl-217707.
Full textAbstract:
This thesis proposes tasks of exercises for mentioned course and design and creation of automated evaluation system for these exercises. This thesis focuses on discussion and exemplary solutions of possible tasks of each exercise and description of created automated evaluation system. For evaluation program are made tests with chosen special data sets, which will prove it’s functionality in general data sets.
38
Tankam, Patrice. "Méthodes d'holographie numérique couleur pour la métrologie sans contact en acoustique et mécanique." Phd thesis, Université du Maine, 2010. http://tel.archives-ouvertes.fr/tel-00647931.
Full textAbstract:
Ce travail de thèse propose des méthodes d'holographie numérique couleur pour la mesure sans contact dans le domaine de l'acoustique, de la mécanique du solide et de la mécanique des fluides. Les méthodes développées pourraient répondre à diverses problématiques académiques et industrielles, notamment, la détection de défauts et de fissurations dans des structures composites, l'analyse des déformations et des vibrations dans leurs 3 composantes, le contrôle de surface et de forme ou l'analyse de fluides dans le cas d'écoulements turbulents à mach subsonique et à mach supersonique. Cette thèse est organisée autour de 5 chapitres qui résument les différents axes de travail. Dans un premier temps, nous avons rappelé quelques aspects théoriques de l'holographie numérique et complété la connaissance du processus de formation des images holographiques numériques en proposant une formulation analytique généralisée incluant les courbures des ondes de référence et de reconstruction numérique, ainsi que des phénomènes non linéaires à l'enregistrement. En particulier nous nous sommes intéressés à développer un modèle de formation des images tenant compte de la saturation des pixels du capteur. Nous avons démontré que, bien que le phénomène de saturation soit non linéaire, nous pouvons néanmoins écrire la relation objet-image avec une formulation analytique linéaire utilisant des produits de convolution. L'analyse théorique a été complétée par une étude expérimentale qui a permis de valider l'approche retenue. Le modèle général de reconstruction permet dans un second temps de définir des stratégies pour la reconstruction numérique d'objets étendus encodés dans des hologrammes couleur. L'algorithme doit avoir pour spécificité de conserver l'horizon reconstruit indépendamment de la longueur d'onde d'enregistrement. Dans ce contexte, ce travail nous a conduit à proposer un algorithme de transformée de Fresnel avec zéro-padding dépendant de la longueur d'onde, puis des algorithmes de convolution à balayage spectral et à grandissement variable qui sont basés sur le concept d'adaptation de bande passante spatiale. Par ailleurs, pour les problématiques rencontrées en mécanique des fluides, nous avons adapté une méthode de filtrage par transformée de Fourier à la reconstruction des objets de phase. Ces algorithmes ont été appliqués expérimentalement à des enregistrements monochromes, bi-couleurs et tri chromatiques montrant ainsi leur grande versatilité. Les deux dernières parties de la thèse traitent du développement de dispositifs holographiques numériques à deux et trois longueurs d'onde pour la métrologie sans contact. Le volet expérimental couvre les champs de la mécanique du solide, l'acoustique et la mécanique des fluides. Nous avons développé plusieurs architectures : architectures bi-couleur et tri-couleur à multiplexage spatial des hologrammes, architecture bi-couleur et tri-chromatique à codage par stack de photodiode, architecture tri chromatique à codage par triple capteur. Les deux dernières architectures s'avèrent les plus simples à mettre en oeuvre. Le banc expérimental bi-couleur à multiplexage spatial a été appliqué à des problématiques industrielles et académiques. La première concerne l'identification de causes de fissuration de capacité sur des composants PCB pour l'automobile. Nous avons mis en évidence une anisotropie de contrainte sur la capacité lors de l'encastrement du PCB dans son support, conduisant ainsi à la probable fissuration de l'élément. La seconde application constitue une première tentative de mise en évidence de modes tourbillonnaires dans des milieux granulaires non consolidés. Nous avons développé deux stratégies de détermination du mouvement 3D du milieu à partir de l'enregistrement d'hologrammes bi-couleurs multiplexés spatialement. Les résultats expérimentaux des modes planaires montrent des tourbillons à certaines fréquences d'excitation du milieu. Nos résultats expérimentaux montrent quelques contradictions dont les causes ont été identifiées. Afin de lever toute ambiguïté nous avons développé un banc d'holographie numérique pseudo-pulsée à 3 couleurs permettant la détermination exclusive de chacune des composantes du vecteur déplacement. Dans le cadre d'une collaboration avec l'ONERA (centre de Lille), nous avons développé une méthode d'holographie numérique tri-chromatique pour l'étude d'écoulements tourbillonnaires. La méthode proposée permet la détermination du champ d'indice ou de masse volumique dans un écoulement et, après validation, a été appliquée à plusieurs cas d'écoulements à mach subsonique et à mach supersonique. Les résultats expérimentaux obtenus sont une première démonstration des potentialités d'application de la méthode d'holographie numérique couleur dans le domaine de la mécanique des fluides.
39
Hanzálek, Pavel. "Praktické ukázky zpracování signálů." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2019. http://www.nusl.cz/ntk/nusl-400849.
Full textAbstract:
The thesis focuses on the issue of signal processing. Using practical examples, it tries to show the use of individual signal processing operations from a practical point of view. For each of the selected signal processing operations, an application is created in MATLAB, including a graphical interface for easier operation. The division of the thesis is such that each chapter is first analyzed from a theoretical point of view, then it is shown using a practical demonstration of what the operation is used in practice. Individual applications are described here, mainly in terms of how they are handled and their possible results. The results of the practical part are presented in the attachment of the thesis.
40
Bodnarova, Adriana. "Texture analysis for automatic visual inspection and flaw detection in textiles." Thesis, Queensland University of Technology, 2000.
Find full text
41
Kato, Fernando Hideki. "Análise de carteiras em tempo discreto." Universidade de São Paulo, 2004. http://www.teses.usp.br/teses/disponiveis/12/12139/tde-24022005-005812/.
Full textAbstract:
Nesta dissertação, o modelo de seleção de carteiras de Markowitz será estendido com uma análise em tempo discreto e hipóteses mais realísticas. Um produto tensorial finito de densidades Erlang será usado para aproximar a densidade de probabilidade multivariada dos retornos discretos uniperiódicos de ativos dependentes. A Erlang é um caso particular da distribuição Gama. Uma mistura finita pode gerar densidades multimodais não-simétricas e o produto tensorial generaliza este conceito para dimensões maiores. Assumindo que a densidade multivariada foi independente e identicamente distribuída (i.i.d.) no passado, a aproximação pode ser calibrada com dados históricos usando o critério da máxima verossimilhança. Este é um problema de otimização em larga escala, mas com uma estrutura especial. Assumindo que esta densidade multivariada será i.i.d. no futuro, então a densidade dos retornos discretos de uma carteira de ativos com pesos não-negativos será uma mistura finita de densidades Erlang. O risco será calculado com a medida Downside Risk, que é convexa para determinados parâmetros, não é baseada em quantis, não causa a subestimação do risco e torna os problemas de otimização uni e multiperiódico convexos. O retorno discreto é uma variável aleatória multiplicativa ao longo do tempo. A distribuição multiperiódica dos retornos discretos de uma seqüência de T carteiras será uma mistura finita de distribuições Meijer G. Após uma mudança na medida de probabilidade para a composta média, é possível calcular o risco e o retorno, que levará à fronteira eficiente multiperiódica, na qual cada ponto representa uma ou mais seqüências ordenadas de T carteiras. As carteiras de cada seqüência devem ser calculadas do futuro para o presente, mantendo o retorno esperado no nível desejado, o qual pode ser função do tempo. Uma estratégia de alocação dinâmica de ativos é refazer os cálculos a cada período, usando as novas informações disponíveis. Se o horizonte de tempo tender a infinito, então a fronteira eficiente, na medida de probabilidade composta média, tenderá a um único ponto, dado pela carteira de Kelly, qualquer que seja a medida de risco. Para selecionar um dentre vários modelos de otimização de carteira, é necessário comparar seus desempenhos relativos. A fronteira eficiente de cada modelo deve ser traçada em seu respectivo gráfico. Como os pesos dos ativos das carteiras sobre estas curvas são conhecidos, é possível traçar todas as curvas em um mesmo gráfico. Para um dado retorno esperado, as carteiras eficientes dos modelos podem ser calculadas, e os retornos realizados e suas diferenças ao longo de um backtest podem ser comparados.In this thesis, Markowitzs portfolio selection model will be extended by means of a discrete time analysis and more realistic hypotheses. A finite tensor product of Erlang densities will be used to approximate the multivariate probability density function of the single-period discrete returns of dependent assets. The Erlang is a particular case of the Gamma distribution. A finite mixture can generate multimodal asymmetric densities and the tensor product generalizes this concept to higher dimensions. Assuming that the multivariate density was independent and identically distributed (i.i.d.) in the past, the approximation can be calibrated with historical data using the maximum likelihood criterion. This is a large-scale optimization problem, but with a special structure. Assuming that this multivariate density will be i.i.d. in the future, then the density of the discrete returns of a portfolio of assets with nonnegative weights will be a finite mixture of Erlang densities. The risk will be calculated with the Downside Risk measure, which is convex for certain parameters, is not based on quantiles, does not cause risk underestimation and makes the single and multiperiod optimization problems convex. The discrete return is a multiplicative random variable along the time. The multiperiod distribution of the discrete returns of a sequence of T portfolios will be a finite mixture of Meijer G distributions. After a change of the distribution to the average compound, it is possible to calculate the risk and the return, which will lead to the multiperiod efficient frontier, where each point represents one or more ordered sequences of T portfolios. The portfolios of each sequence must be calculated from the future to the present, keeping the expected return at the desired level, which can be a function of time. A dynamic asset allocation strategy is to redo the calculations at each period, using new available information. If the time horizon tends to infinite, then the efficient frontier, in the average compound probability measure, will tend to only one point, given by the Kellys portfolio, whatever the risk measure is. To select one among several portfolio optimization models, it is necessary to compare their relative performances. The efficient frontier of each model must be plotted in its respective graph. As the weights of the assets of the portfolios on these curves are known, it is possible to plot all curves in the same graph. For a given expected return, the efficient portfolios of the models can be calculated, and the realized returns and their differences along a backtest can be compared.
42
Lin, Chin-Tung, and 林鑫彤. "InverseMV: Composing Piano Scores with a Convolutional Video-Music Transformer." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/y27qbg.
Full textAbstract:
碩士國立政治大學
資訊科學系
107
With the wide popularity of social media including Facebook, Twitter, Instagram, YouTube, etc. and the modernization of mobile photography, users on social media tend to watch and send videos rather than text. People want their video with a high click-through rate. However, such video requires great editing skill and perfect matching music, which are very difficult for common people. On top of that, people creating soundtrack suffer from the lack of ownership of musical pieces. The music generated from a model instead of existing music conduces to preventing from breaching copyright. The rise of deep learning brought out much work using a model based on the neural network to generate symbolic music. However, to the best of our knowledge, there is no work trying to compose music for video and no dataset with paired video and music. Therefore, we release a new dataset composed of over 7 hours of piano scores with fine alignment between pop music videos and midi files. We propose a model VMT(Video-Music Transformer) that generates piano scores from video frames, and then evaluate our model with seq2seq and obtain better music smooth and relevance of video.
43
JIN, YI-MING, and 金益明. "Convolution-based algorithms for discrete sinusoidal transforms." Thesis, 1991. http://ndltd.ncl.edu.tw/handle/00412644879232058537.
Full text
44
Lin, Tzu-Chia, and 林子嘉. "Automated Arrhythmia Detection using Hilbert-Huang Transform Based Convolutional Neural Network." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/7zyxue.
Full textAbstract:
碩士國立中央大學
資訊工程學系
107
In this thesis, a novel approach to arrhythmia-based signal classification is introduced. The objective is to properly identify three classes of patients exhibiting normal sinus rhythm, atrial fibrillation, and other rhythm. The proposed method apply Hilbert-Huang transform on raw signal to generate noise-free reconstruction of the original containing temporal variations as input for classification mechanism to learn representative features. The features are directly learned by a computer vision technique known as Convolutional Neural Network, thus replacing traditional methods of relying on experts to handcraft features. To summarize, this thesis contains two major processes: utilize a nonlinear and non-stationary signal processing technique to produce input, and to feed reconstructed signal containing representative features to CNN for multi-classification task. The experimental results indicate the effectiveness of this method, removing the need of human involvement in the process of feature selection. Through analyses and stimulations, the effectiveness of the proposed ECG-classification method is evaluated.
45
Chiao-Yuan, Jeng, and 鄭朝元. "Multidimension Generalized Discrete Hartley Transform and related convolution." Thesis, 1995. http://ndltd.ncl.edu.tw/handle/88133818661332311284.
Full text
46
Hlawitschka, Mario, Julia Ebling, and Gerik Scheuermann. "Convolution and Fourier Transform of Second Order Tensor Fields." 2004. https://ul.qucosa.de/id/qucosa%3A32974.
Full textAbstract:
The goal of this paper is to transfer convolution, correlation and Fourier transform to second order tensor fields. Convolution of two tensor fields is defined using matrix multiplication. Convolution of a tensor field with a scalar mask can thus be described by multiplying the scalars with the real unit matrix. The Fourier transform of tensor fields defined in this paper corresponds to Fourier transform of each of
the tensor components in the field. It is shown that for this convolution and Fourier transform, the well known convolution theorem holds and optimization in speed can be achieved by using Fast Fourier transform algorithms. Furthermore, pattern matching on tensor fields based on this convolution is described.
47
杜維洲. "Novel Convolution-Based computations of the discrete cosine transform." Thesis, 1992. http://ndltd.ncl.edu.tw/handle/63817327093807243898.
Full text
48
Chen, Yen-Feng, and 陳彥灃. "Facial Gender Classification based on Convolutional Neural Networks Cascaded with Discrete Wavelet Transform." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/05937291010012756825.
Full textAbstract:
碩士國立暨南國際大學
電機工程學系
105
Facial gender classification based on deep convolutional neural networks (DCNNs) cascaded with discrete wavelet transform (DWT) is a work in biometrics by means of the deep learning (DL) architecture. The applied DL architecture is convolutional neural networks (CNNs) which is appropriate to image processing. Biometrics is a subject of pattern recognition while deep learning is a novel machine learning architecture, which is considered the most powerful classifier among the state-of-the-art. The architecture of this work is that three specified channels of CNNs connected in parallel are cascaded with three frequency bands of DWT, respectively. The DWT process extracts critical features from input faces by which the number of kernel of the CNNs can be reduced. This will reduce the training time and the impact of random. Adience 3D dataset is applied as the benchmark.
49
Keating, Geoffrey. "Fast fourier transforms and the Fujitsu VPP300." Phd thesis, 2000. http://hdl.handle.net/1885/147291.
Full text
50
呂芳發. "Two dimension generalized discrete hartley transforms fast algorithm and related convolutions." Thesis, 1992. http://ndltd.ncl.edu.tw/handle/15945666499828844959.
Full text