Dissertations / Theses on the topic 'Zernike moments'

L'expérience HESS consiste en un système de télescopes permettant d'observer les rayonnements cosmiques. Compte tenu des résultats majeurs obtenus depuis son installation, la seconde phase du projet a été engagée. Celle-ci passe par l'ajout d'un télescope plus sensible et plus grand que ses prédécesseurs. Toutes les données collectées par ce télescope ne pouvant être conservées à cause des limites de stockage, un système de déclencheur (trigger) performant doit être mis en place. L'objectif de cette thèse est de proposer une solution de reconnaissance de formes en temps réel, embarquée sur le télescope. La première partie de la thèse a consisté à élaborer une chaine de reconnaissance des formes pour ce trigger. Une chaine de traitement à base de réseau de neurones et des moments de Zernike a été validée. La seconde partie de la thèse a porté sur l'implantation des algorithmes retenus sur FPGA en tenant compte des contraintes en termes de ressources et de temps d'exécution
The HESS experiment consists of a system of telescopes destined to observe cosmic rays. Since the project has achieved a high level of performances, a second phase of the project has been initiated. This implies the addition of a new telescope which is capable of collecting a huge amount of images. As all data collected by the telescope can not be retained because of storage limitations, a new real-time system trigger must be designed in order to select interesting events on the fly. The purpose of this thesis was to propose a trigger solution to efficiently discriminate events captured by the telescope. The first part of this thesis was to develop pattern recognition algorithms to be implemented within the trigger. A processing chain based on neural networks and Zernike moments has been validated. The second part of the thesis has focused on the implementation of the proposed algorithms onto FPGA, taking into account the application constraints in terms of resources and execution time

The 3-D fluorescence microscope has become the method of choice in biological sciences for living cells study. However, the data acquired with conventional3-D fluorescence microscopy are not quantitatively significant because of distortions induced by the optical acquisition process. Reliable measurements need the correction of theses distortions. Knowing the instrument impulse response, also known as the PSF, one can consider the backward process of convolution induced by the microscope, known as "deconvolution". However, when the system response is not invariant in the observation field, the classical algorithms can introduce large errors in the results. In this thesis we propose a new approach, which can be easily adapted to any classical deconvolution algorithm, direct or iterative, for bypassing the non-invariance PSF problem, without any modification to the later. Based on the hypothesis that the minimal error in a restored image using non-invariance assumption is located near the used PSF position, the EMMA (Evolutive Merging Masks Algorithm) blends multiple deconvolutions in the invariance assumption using a specific merging mask set. In order to obtain sufficient number of measured PSF at various depths for a better restoration using EMMA (or any other depth-variant deconvolution algorithm) we propose a 3D PSF interpolation algorithm based on the image moments theory using Zernike polynomials as decomposition base. The known PSF are decomposed into Zernike moments set and each moment's variation is fitted into a polynomial function, the resulting functions are then used to interpolate the needed PSF's Zernike moments set to reconstruct the interpolated PSF.

This thesis presents a novel lip-reading approach to classifying utterances from video data, without evaluating voice signals. This work addresses two important issues which are • the efficient representation of mouth movement for visual speech recognition • the temporal segmentation of utterances from video. The first part of the thesis describes a robust movement-based technique used to identify mouth movement patterns while uttering phonemes. This method temporally integrates the video data of each phoneme into a 2-D grayscale image named as a motion template (MT). This is a view-based approach that implicitly encodes the temporal component of an image sequence into a scalar-valued MT. The data size was reduced by extracting image descriptors such as Zernike moments (ZM) and discrete cosine transform (DCT) coefficients from MT. Support vector machine (SVM) and hidden Markov model (HMM) were used to classify the feature descriptors. A video speech corpus of 2800 utterances was collected for evaluating the efficacy of MT for lip-reading. The experimental results demonstrate the promising performance of MT in mouth movement representation. The advantages and limitations of MT for visual speech recognition were identified and validated through experiments. A comparison between ZM and DCT features indicates that th e accuracy of classification for both methods is very comparable when there is no relative motion between the camera and the mouth. Nevertheless, ZM is resilient to rotation of the camera and continues to give good results despite rotation but DCT is sensitive to rotation. DCT features are demonstrated to have better tolerance to image noise than ZM. The results also demonstrate a slight improvement of 5% using SVM as compared to HMM. The second part of this thesis describes a video-based, temporal segmentation framework to detect key frames corresponding to the start and stop of utterances from an image sequence, without using the acoustic signals. This segmentation technique integrates mouth movement and appearance information. The efficacy of this technique was tested through experimental evaluation and satisfactory performance was achieved. This segmentation method has been demonstrated to perform efficiently for utterances separated with short pauses. Potential applications for lip-reading technologies include human computer interface (HCI) for mobility-impaired users, defense applications that require voice-less communication, lip-reading mobile phones, in-vehicle systems, and improvement of speech-based computer control in noisy environments.

Made available in DSpace on 2015-04-09T12:28:28Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-10-24Bitstream added on 2015-04-09T12:47:21Z : No. of bitstreams: 1 000812794.pdf: 1525344 bytes, checksum: b68f6da113153c038916e9bd3f57c375 (MD5)
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Nesta pesquisa foi estudado um m etodo de reconhecimento de contornos de telhado de edif cios em imagens digitais de alta resolu c~ao, que classi ca-os com rela c~ao a sua forma. O m etodo baseia-se nos momentos de Zernike, que s~ao baseados nos polin omios ortogonais de Zernike, em que cria-se um vetor de caracter sticas para cada regi~ao da imagem, que deve ser previamente segmentada de maneira que seus objetos sejam divididos em diferentes regi~oes. Este m etodo para a descri c~ao de forma baseia-se na area do objeto de interesse e possui a caracter stica dos momentos serem invariantes em rela c~ao as transforma c~oes geom etricas de rota c~ao, transla c~ao e escala, que o torna atrativo para o problema de an alise de imagem proposto. Desse modo, foi criada uma base de dados contendo esbo cos (ou modelos) de poss veis apari c~oes de contornos de telhado de edif cio numa dada cena, para que seja associado tamb em um vetor de caracter sticas de Zernike para estes esbo cos. Assim, a dist ancia euclidiana entre este vetor e o vetor de caracter sticas calculado a partir de uma regi~ao segmentada na imagem, permite dizer se a regi~ao dada corresponde a um contorno de edif cio ou a outro objeto. A capacidade de discrimina c~ao do m etodo proposto entre diferentes formas de edif cios, e tamb em entre formas de edif cios e n~ao edif cios foi avaliada experimentalmente e mostrou resultados positivos.
In this research, a method of recognition of building roof contours in high-resolution digital images which classi es them with respect to their form was studied. The method is based on Zernike moments, which are based on orthogonal Zernike polynomials and it creates a feature vector for each image region. The image segmentation has to be made rst to de ne di erent regions for its objects. This method for shape analysis is based on the object area of interest and the moments has the characteristic of being invariant under geometric transformations of rotation, translation and scaling, this makes it attractive to the proposed image analysis problem. Thus, a database containing sketches (or models) of possible appearances of building roof contours in a given scene was created, so a Zernike feature vector was also associated for these sketches. Therefore, the Euclidean distance between this vector and the feature vector calculated from a segmented region in the image lets say if the given region corresponds to a building contour or other object. The capacity of the proposed method in discriminating di erent building shapes and also in discriminating building shapes from non-building shapes was evaluated experimentally and it showed positive results.

Orientador: Aluir Porfírio Dal Poz
Banca: Edson Aparecido Mitishita
Banca: Aylton Pagamisse
Resumo: Nesta pesquisa foi estudado um m etodo de reconhecimento de contornos de telhado de edif cios em imagens digitais de alta resolu c~ao, que classi ca-os com rela c~ao a sua forma. O m etodo baseia-se nos momentos de Zernike, que s~ao baseados nos polin^omios ortogonais de Zernike, em que cria-se um vetor de caracter sticas para cada regi~ao da imagem, que deve ser previamente segmentada de maneira que seus objetos sejam divididos em diferentes regi~oes. Este m etodo para a descri c~ao de forma baseia-se na area do objeto de interesse e possui a caracter stica dos momentos serem invariantes em rela c~ao as transforma c~oes geom etricas de rota c~ao, transla c~ao e escala, que o torna atrativo para o problema de an alise de imagem proposto. Desse modo, foi criada uma base de dados contendo esbo cos (ou modelos) de poss veis apari c~oes de contornos de telhado de edif cio numa dada cena, para que seja associado tamb em um vetor de caracter sticas de Zernike para estes esbo cos. Assim, a dist^ancia euclidiana entre este vetor e o vetor de caracter sticas calculado a partir de uma regi~ao segmentada na imagem, permite dizer se a regi~ao dada corresponde a um contorno de edif cio ou a outro objeto. A capacidade de discrimina c~ao do m etodo proposto entre diferentes formas de edif cios, e tamb em entre formas de edif cios e n~ao edif cios foi avaliada experimentalmente e mostrou resultados positivos.
Abstract: In this research, a method of recognition of building roof contours in high-resolution digital images which classi es them with respect to their form was studied. The method is based on Zernike moments, which are based on orthogonal Zernike polynomials and it creates a feature vector for each image region. The image segmentation has to be made rst to de ne di erent regions for its objects. This method for shape analysis is based on the object area of interest and the moments has the characteristic of being invariant under geometric transformations of rotation, translation and scaling, this makes it attractive to the proposed image analysis problem. Thus, a database containing sketches (or models) of possible appearances of building roof contours in a given scene was created, so a Zernike feature vector was also associated for these sketches. Therefore, the Euclidean distance between this vector and the feature vector calculated from a segmented region in the image lets say if the given region corresponds to a building contour or other object. The capacity of the proposed method in discriminating di erent building shapes and also in discriminating building shapes from non-building shapes was evaluated experimentally and it showed positive results.
Mestre

Submitted by Mayara Nascimento (mayara.nascimento@ufba.br) on 2016-05-31T14:15:01Z No. of bitstreams: 1 Igor Bastos - Dissertação VFinal.pdf: 3703212 bytes, checksum: a00013910865dacb8025d56659076abb (MD5)
Approved for entry into archive by Alda Lima da Silva (sivalda@ufba.br) on 2016-06-03T23:17:29Z (GMT) No. of bitstreams: 1 Igor Bastos - Dissertação VFinal.pdf: 3703212 bytes, checksum: a00013910865dacb8025d56659076abb (MD5)
Made available in DSpace on 2016-06-03T23:17:29Z (GMT). No. of bitstreams: 1 Igor Bastos - Dissertação VFinal.pdf: 3703212 bytes, checksum: a00013910865dacb8025d56659076abb (MD5)
Gestos são ações corporais não-verbais voltadas para a expressão de algum significado. Estes incluem movimentos de mãos, face, braços, dedos, entre outros, sendo abordados por trabalhos que visam reconhecê-los para promover interações humanas com sistemas computacionais. Devido à grande aplicabilidade do reconhecimento de gestos, tem-se notado que estes trabalhos estão se tornando mais comuns, utilizando técnicas e metodologias mais elaboradas e capazes de prover resultados cada vez melhores. A opção por quais técnicas aplicar para o reconhecimento de gestos varia de acordo com a estratégia empregada em cada trabalho e quais aspectos são utilizados para este reconhecimento. Tem-se, por exemplo, trabalhos baseados no uso de modelos estatísticos. Outros optam pela aquisição de características geométricas de mãos e partes do corpo, enquanto outros, dentre os quais se enquadra o presente trabalho, optam pelo uso de descritores e classificadores, responsáveis por extrair características das imagens relevantes para o seu reconhecimento e; por realizar a classificação efetiva dos gestos baseado nestas informações. Neste âmbito, o presente trabalho visa elaborar, aplicar e apresentar uma abordagem para o reconhecimento de gestos, embasando-se em uma revisão da literatura a respeito das principais técnicas e metodologias empregadas para este fim e escolhendo como campo prático, a Língua Brasileira de Sinais (Libras). Para a extração de informações das imagens, optou-se pelo uso de um vetor de características resultante da aplicação dos descritores Histograma de Gradientes Orientados (HOG) e Momentos Invariantes de Zernike (MIZ), os quais voltam-se para as formas e contornos presentes nas imagens. Para o reconhecimento, foi utilizado o classificador Perceptron Multicamada, sendo este disposto em uma arquitetura onde o processo de classificação é dividido em 2 estágios. Devido à inexistência de datasets públicos da Libras, fez-se necessária, com o auxílio de especialistas da língua e alunos surdos, a criação de um dataset de 9600 imagens, as quais referem-se a 40 sinais da Libras. Isso fez com que a presente abordagem partisse desta criação do dataset até a etapa final de classificação dos sinais. Por fim, testes foram realizados e obteve-se 96,77% de taxa de acerto, evidenciando um alto índice de acerto. Este resultado foi validado considerando possíveis ameaças à abordagem, como a realização de testes considerando um indivíduo não-presente no conjunto de treinamento do classificador e a aplicação da abordagem em um dataset público de gestos.

L'expérience HESS consiste en un système de télescopes permettant d'observer les rayonnements cosmiques. Compte tenu des résultats majeurs obtenus depuis son installation, la seconde phase du projet a été engagée. Celle-ci est en cours de réalisation et passe par l'ajout d'un télescope plus sensible et plus grand que ses prédécesseurs. Toutes les données collectées par ce télescope ne peuvent pas être conservées à cause des limites de stockage. Par conséquent, un système de déclencheur, dit trigger, performant doit être mis en place. L'objectif de cette thèse est de proposer une solution de reconnaissance de formes en temps réel dans un contexte fortement contraint et qui sera embarquée sur le télescope. La première partie de la thèse a consisté à élaborer une chaîne de reconnaissance des formes pour ce trigger. Une chaîne de traitement à base de réseau de neurones et des moments de Zernike a été validée. La seconde partie de la thèse a porté sur l'implantation des algorithmes retenus sur une cible FPGA en tenant compte des contraintes en termes de ressources et de temps d'exécution.

碩士
國立交通大學
電機與控制工程系
89
An estimation approach of image rotation angle is proposed here. A basic assumption that the center of image rotation must be known is made, so that we can estimate the precise image rotation angles from the comparison between the rotated images and the reference one. Furthermore, the Zernike moment algorithm with subsampling and interpolation techniques can increase the image resolution and compute Zernike moments more accurately on the premise that keep the image acquiring hardware instrument unchanged. In addition, a K-means clustering algorithm is used to classify valuable entries from all the rotation angle candidates extracted via Zernike moments and sum up them by weighting factors based on image reconstruction. After that, the estimation of image rotation angle with high resolution is derived. Two experiments are reported which are on rotated images generated by computer and captured by CCD camera, respectively. The experimental results show that the proposed method has good performance for image rotation angle estimation.

Zernike moments are sets of mathematical quantities that uniquely characterize an image. It is known that they are invariant under rotation and reflection and robust to noise. In this thesis several other algorithms have been used to calculate these moments. The intent of this thesis is: 1. to use the classical method and the algorithms to reconstruct an image using Zernike moments and study their accuracy and 2. to examine if the invariance and noise insensitivity property of the calculated Zernike moments are upheld by these procedures. It is found that the constructed images using these algorithms do not resemble the original image. This prevents us from carrying out further study of these algorithms. The classical method has been successfully used to reconstruct an image when the height and width are equal. The classical method is also shown to be invariant under rotation and reflection and robust to Poisson noise. xxxvii

The field of face recognition has been explored a lot and the work is still going on. In the presented work we have proposed a novel approach for face recognition using moments. Four methods have been used for feature extraction: Hu moments, Zernike moments, Legendre moments and Cumulants. Hu moments are a set of seven moments which have been derived from the conventional geometric moments. These are invariant against rotation, scaling and translation. Legendre moments and Zernike moments have an orthogonal basis set and can be used to represent an image with a minimum amount of information redundancy. They are based on the theory of orthogonal polynomials and can be used to recover an image from moment invariants. Cumulants are sensitive to the image details and therefore are suitable for representing the features of images. For feature extraction, moments of different orders are calculated which form the feature vectors. The obtained feature vectors are stored in the database and are classified using three classifiers: Minimum Distance Classifier, Support Vector Machine and K Nearest Neighbor. For testing the proposed approach, the ORL (Olivetty Research Laboratories) database is used. It consists of 40 subjects, each having 10 orientations.

This thesis presents a technique for face recognition that is based on image registration. The image registration technique is based on finding a set of feature points in the two images and using these feature points for registration. This is done in four steps. In the first, images are filtered with the Mexican hat wavelet to obtain the feature point locations. In the second, the Zernike moments of neighbourhoods around the feature points are calculated and compared in the third step to establish correspondence between feature points in the two images and in the fourth the transformation parameters between images are obtained using an iterative weighted least squares technique. The face recognition technique consists of three parts, a training part, an image registration part and a post-processing part. During training a set of images are chosen as the training images and the Zernike moments for the feature points of the training images are obtained and stored. In the registration part, the transformation parameters to register the training images with the images under consideration are obtained. In the post-processing, these transformation parameters are used to determine whether a valid match is found or not. The performance of the proposed method is evaluated using various face databases and it is compared with the performance of existing techniques. Results indicate that the proposed technique gives excellent results for face recognition in conditions of varying pose, illumination, background and scale. These results are comparable to other well known face recognition techniques.

碩士
國立宜蘭大學
電子工程學系碩士班
99
There are many face recognition methods such as principal component analysis (PCA), neural network and wavelet transform etc. recently. Conventional methods to deal with color face image are transformed to gray or binary image, which may loss some significant color features lead to accuracy rate descending. In order to preserve the image color information we use quaternion Zernike moment invariant extracted feature that directly process in color space and with the naïve Bayes classifier recognition. Two database we used in this paper, they are IMM database and face94 database to achieve face recognition. At first a preprocessing for color face images is performed, including pixel normalization, image resizing and color space transform. Second, the quaternion Zernike moment invariants are used to extract face image feature. Finally the naïve Bayes classifier is used to recognition face image and we also compare the results with other classifiers. Keyword：color face recognition, quaternion Zernike moment invariants

博士
國立交通大學
資訊科學與工程研究所
97
In 3D computer vision a scene in the real world is represented by multiple views imaged under different viewpoints and illumination conditions. The spatial and temporal relationships across these views are important to scene analysis and understanding. To derive these relationships the global and local features of the objects (foreground and background) in the scene are the clues. The local features related to the local object surface patches or regions are more robust to viewpoint change than the global features. In addition, the invariance under the photometric transformations such as blur, illumination, scale, noise, JPEG compression is also receiving great attention. In this dissertation subjects related to the local image representation, matching, and recognition under the above image variations are addressed. First, a new distinctive image descriptor to represent the normalized regions extracted by an affine region detector is proposed which primarily comprises the Zernike moment (ZM) phase information. An accurate and robust estimation of a possible rotation angle between a pair of normalized regions is then described, which will be used to measure the similarity between two matching regions. The discriminative power of the new ZM phase descriptor is compared with five major existing region descriptors based on the precision-recall criterion. The experimental results involving more than 15 million region pairs indicate the proposed ZM phase descriptor has, overall speaking, the best performance under the common photometric and geometric transformations. Both quantitative and qualitative analyses on the descriptor performances are given to account for the performance discrepancy. Second, the proposed ZM phase descriptor is further extended to present a new recognition method of logos imaged by mobile phone cameras. The logo recognition can be incorporated with mobile phone services for use in enterprise identification, corporate website access, traffic sign reading, security check, content awareness, and the related applications. The main challenge to applying the logo recognition for mobile phone applications is the inevitable photometric and geometric transformations. The proposed ZM phase recognition method is associated with two similarity measures. The logo classification and retrieval experimental results show that the proposed ZM phase method has the best performance under the typical photometric and geometric transformations, compared with other three major existing methods. Finally, as for the one-to-one feature matching correspondences in view registration, we propose an efficient registration method different from the traditional methods. We take advantage of preprocessing of the reference image offline to gather the important statistics for guiding image registration. That is, we introduce five planning strategies to sort the feature points in the reference image based on the concepts of (1) feature invariance to image deformation, (2) image noise resistance, (3) distinctive description power, (4) model estimation effectiveness, and (5) partial image overlapping handling capability. Thus, a reference matching database is constructed offline using the above five planning strategies. Then, an online registration process is presented to estimate the transformation model to overlay the reference image over an incoming sensed image. In this way, better registration efficiency can be achieved.

碩士
國立交通大學
控制工程系
84
We propose an object recognition approach that is invariant to translation, scale, and rotation. We assume objects are placed on a uniform background so that we can easily distinguish between the object and its background, and use localized Zernike moments to extract object invariant features. This method not only retains the rotation invariance of Zernike moments, but also extractes the geometrical features of objects. That is, localized Zernike moments can obtain more information from the same image than Zernike moments provide. A supervised fuzzy adaptive Hamming net is equivalent to a fuzzy ARTMAP when used as a classifier, and doesn't need to search when learning and recognition, i.e., it is more efficient than fuzzy ARTMAP. So, we use it as our classifier. Two experiments are reported, 2-D key-pattern recognition and 3-D target recognition. The experimental results show that the proposed method obtained good performance for both 2-D and 3-D object recognition.

碩士
國立暨南國際大學
電機工程學系
98
With an increasing emphasis on security, personal authentication based on biometrics has been receiving extensive attention over the past decade. Among many different biometric technologies, this thesis examines palm-print and hand-shape technique for personal identification and develops a good performance recognition system based on human hand features. It is implemented and tested on VIP-CC Lab. hand image database. The proposed system includes four modules: image acquisition, image pre-processing, feature extraction, and recognition modules. First, the system captures a hand image using digital camera, then uses some image processing algorithms to localize the region of the interest of palm-print and hand-geometry from the hand image via image pre-processing module. Via feature extraction module, in the palmprint and hand-shape, adopts the fast compute Zernike Moment. Experimental results show that the system has an encouraging performance on the VIP-CC Lab. database（including 160 images from 20 classes）. In this Biometric system, we take palmprint and hand-shape by using fast computing Zernike moment. And we get satisfactory results. The system adopts fast computing Zernike moment to extract palm-print and hand-shape features.

碩士
國立中山大學
機械與機電工程學系研究所
94
The need of finding a way to design a company trademark, without the worry of possible infringement on the intellectual property rights, has become exceedingly important as the economy and the accompanying intellectual property concerns advanced greatly in recent years. Traditionally, registered trademarks are stored in image databases and are categorized and retrieved by descriptions and keywords given by human workers. This is extremely time-consuming and considered by many as inappropriate. In this work we focus on image feature and content related techniques, or content-based image retrieval (CBIR) methods. Nevertheless, we still need human inputs since by law the most crucial basis for discerning the similarity or difference of two trademarks has to rely on human’s naked eye. Therefore in this work we created a program which incorporates an man-machine interface allowing users to input various weighting factors each emphasizing a specific feature or shape of the trademark. The Zernike moments, and some new image compactness indices are used in the computations for image comparisons.

碩士
國立清華大學
資訊工程學系
95
This paper present a novel content-based image authentication framework which embeds the semi-fragile image feature into the host image based on wavelet transform. In this framework, two features of a target image from the low frequency domain to generate two watermarks: Zernike moments for classifying of the intentional content modification and sobel edge for indicating the modified location. In particular, we design a systematic method for automatic order selection of Zernike moments and in order to tell if the procession on the image is malicious or not, we also propose a weighted Euclidean distance by its reconstruction process. An important advantage of our approach is that it can tolerate compression and noise to a certain extent while rejecting common tampering to the image like rotation. Experimental results show that the framework can locate the malicious tamper locally, the unit of detection region is 8x8 block, while highly robust to content preserved processing, such as JPEG compression Q>=30 and Gaussian noise variance<=20.