Texture segmentation is an effortless process in scene analysis, yet its mechanisms have not been sufficiently understood. Several theories and algorithms exist for texture discrimination based on vision. These models diverge from one another in algorithmic approaches to address texture imagery using spatial elements and their statistics. Even though there are differences among these approaches, they all begin from the assumption that texture segmentation is a visual task. However, considering that texture is basically a surface property, this assumption can at times be misleading. An interesting possibility is that since surface properties are most immediately accessible to touch, texture perception may be more intimately associated with texture than with vision (it is known that tactile input can affect vision). Coincidentally, the basic organization of the touch (somatosensory) system bears some analogy to that of the visual system. In particular, recent neurophysiological findings showed that receptive fields for touch resemble that of vision, albeit with some subtle differences. The main novelty and contribution of this thesis is in the use of tactile receptive field responses for texture segmentation. Furthermore, we showed that touch-based representation is superior to its vision-based counterpart when used in texture boundary detection. Tactile representations were also found to be more discriminable (LDA and ANOVA). We expect our results to help better understand the nature of texture perception and build more powerful texture processing algorithms. The results suggest that touch has an advantage over vision in texture processing. Findings in this study are expected to shed new light on the role of tactile perception of texture and its interaction with vision, and help develop more powerful, biologically inspired texture segmentation algorithms.

Texture boundary detection (or segmentation) is an important capability of the human visual system. Usually, texture segmentation is viewed as a 2D problem, as the definition of the problem itself assumes a 2D substrate. However, an interesting hypothesis emerges when we ask a question regarding the nature of textures: What are textures, and why did the ability to discriminate texture evolve or develop? A possible answer to this question is that textures naturally define physically distinct surfaces or objects, thus, we can hypothesize that 2D texture segmentation may be an outgrowth of the ability to discriminate surfaces in 3D. In this thesis, I investigated the relative difficulty of learning to segment textures in 2D vs. 3D configurations. It turns out that learning is faster and more accurate in 3D, very much in line with what was expected. Furthermore, I have shown that the learned ability to segment texture in 3D transfers well into 2D texture segmentation, but not the other way around, bolstering the initial hypothesis, and providing an alternative approach to the texture segmentation problem.

Texture boundary detection (or segmentation) is an important capability of the human visual system. Usually, texture segmentation is viewed as a 2D problem, as the definition of the problem itself assumes a 2D substrate. However, an interesting hypothesis emerges when we ask a question regarding the nature of textures: What are textures, and why did the ability to discriminate texture evolve or develop? A possible answer to this question is that textures naturally define physically distinct surfaces or objects, thus, we can hypothesize that 2D texture segmentation may be an outgrowth of the ability to discriminate surfaces in 3D. In this thesis, I investigated the relative difficulty of learning to segment textures in 2D vs. 3D configurations. It turns out that learning is faster and more accurate in 3D, very much in line with what was expected. Furthermore, I have shown that the learned ability to segment texture in 3D transfers well into 2D texture segmentation, but not the other way around, bolstering the initial hypothesis, and providing an alternative approach to the texture segmentation problem.

Texture Segmentation is one of the most challenging areas of computer vision. One reason for this difficulty is the huge variety and variability of textures occurring in real world, making it very difficult to quantitatively study textures. One of the key tools used for texture segmentation is local invariant descriptors. Texture consists of textons, the basic building block of textures, that may vary by small nuisances like illumination variation, deformations, and noise. Local invariant descriptors are robust to these nuisances making them beneficial for texture segmentation. However, grouping dense descriptors directly for segmentation presents a problem: existing descriptors aggregate data from neighborhoods that may contain different textured regions, making descriptors from these neighborhoods difficult to group, leading to significant errors in segmentation. This work addresses this issue by proposing dense local descriptors, called Shape-Tailored Features, which are tailored to an arbitrarily shaped region, aggregating data only within the region of interest. Since the segmentation, i.e., the regions, are not known a-priori, we propose a joint problem for Shape-Tailored Features and the regions. We present a framework based on variational methods. Extensive experiments on a new large texture dataset, which we introduce, show that the joint approach with Shape-Tailored Features leads to better segmentations over the non-joint non Shape-Tailored approach, and the method out-performs existing state-of-the-art.

The contourlet transform is an emerging multiscale multidirection image processing technique. It effectively represents smooth curvature details typical of natural images, overcoming a major drawback of the 2-D wavelet transform. To further exploit its potential, in this research, a statistical model, the contourlet contextual hidden Markov model (C-CHMM), has been developed to characterize contourlet images. A systematic mutual information based context construction procedure has been developed to form an appropriate context for the model. With this contourlet image model, a multiscale segmentation method has also been established for the application to texture images. The segmentation method combines a model comparison approach with a multiscale fusion and a multi-neighbor combination process. It also features a neighborhood selection scheme based on a smoothed context map, for both the model estimation and the neighbor combination. The effectiveness of the image model has been verified through a series of denoising and segmentation experiments. As demonstrated with the denoising performance, this new model for contourlet images is more promising than the state of the art, the contourlet hidden Markov tree (C-HMT) model. The other model being compared with in this work is the wavelet contextual hidden Markov model (W-CHMM). Through the denoising experiments, the presented C-CHMM shows better robustness against noise than the W-CHMM. Moreover, the new model demonstrates its superiority to the wavelet model in the segmentation performance. Through the segmentation experiments, the value of the systematic context construction procedure has been proven. The C-CHMM based segmentation method has also been validated. In comparison with the state of the art methods for the same type, the presented technique shows improved accuracy in segmenting texture patterns of diversified nature. This success in segmentation has further manifested the potential of the newly developed contourlet image model.

3D city models have been widely used in various applications such as urban planning, traffic control, disaster management etc. Efficient visualisation of 3D city models in different levels of detail (LODs) is one of the pivotal technologies to support these applications. In this thesis, a framework is proposed to visualise the 3D city models online. Then, generalisation methods are studied and tailored to create 3D city scenes in different scales dynamically. Multiple representation structures are designed to preserve the generalisation results on different level. Finally, the quality of the generalised 3D city models is evaluated by measuring the visual similarity with the original models.   In the proposed online visualisation framework, City Geography Makeup Language (CityGML) is used to represent city models, then 3D scenes in Extensible 3D (X3D) are generated from the CityGML data and dynamically updated to the user side for visualisation in the Web-based Graphics Library (WebGL) supported browsers with X3D Document Object Model (X3DOM) technique. The proposed framework can be implemented at the mainstream browsers without specific plugins, but it can only support online 3D city model visualisation in small area. For visualisation of large data volumes, generalisation methods and multiple representation structures are required.   To reduce the 3D data volume, various generalisation methods are investigated to increase the visualisation efficiency. On the city block level, the aggregation and typification methods are improved to simplify the 3D city models. On the street level, buildings are selected according to their visual importance and the results are stored in the indexes for dynamic visualisation. On the building level, a new LOD, shell model, is introduced. It is the exterior shell of LOD3 model, in which the objects such as windows, doors and smaller facilities are projected onto walls.  On the facade level, especially for textured 3D buildings, image processing and analysis methods are employed to compress the texture.   After the generalisation processes on different levels, multiple representation data structures are required to store the generalised models for dynamic visualisation. On the city block level the CityTree, a novel structure to represent group of buildings, is tested for building aggregation. According to the results, the generalised 3D city model creation time is reduced by more than 50% by using the CityTree. Meanwhile, a Minimum Spanning Tree (MST) is employed to detect the linear building group structures in the city models and they are typified with different strategies. On the building level and the street level, the visible building index is created along the road to support building selection. On facade level the TextureTree, a structure to represent building facade texture, is created based on the texture segmentation.   Different generalisation strategies lead to different outcomes. It is critical to evaluate the quality of the generalised models. Visually salient features of the textured building models such as size, colour, height, etc. are employed to calculate the visual difference between the original and the generalised models. Visual similarity is the criterion in the street view level building selection. In this thesis, the visual similarity is evaluated locally and globally. On the local level, the projection area and the colour difference between the original and the generalised models are considered. On the global level, the visual features of the 3D city models are represented by Attributed Relation Graphs (ARG) and their similarity distances are calculated with the Nested Earth Mover’s Distance (NEMD) algorithm.   The overall contribution of this thesis is that 3D city models are generalised in different scales (block, street, building and facade) and the results are stored in multiple representation structures for efficient dynamic visualisation, especially for online visualisation. / QC 20111116 / ViSuCity

Stimuli outside classical receptive fields have been shown to exert significant influence over the activities of neurons in primary visual cortexWe propose that contextual influences are used for pre-attentive visual segmentation, in a new framework called segmentation without classification. This means that segmentation of an image into regions occurs without classification of features within a region or comparison of features between regions. This segmentation framework is simpler than previous computational approaches, making it implementable by V1 mechanisms, though higher leve l visual mechanisms are needed to refine its output. However, it easily handles a class of segmentation problems that are tricky in conventional methods. The cortex computes global region boundaries by detecting the breakdown of homogeneity or translation invariance in the input, using local intra-cortical interactions mediated by the horizontal connections. The difference between contextual influences near and far from region boundaries makes neural activities near region boundaries higher than elsewhere, making boundaries more salient for perceptual pop-out. This proposal is implemented in a biologically based model of V1, and demonstrated using examples of texture segmentation and figure-ground segregation. The model performs segmentation in exactly the same neural circuit that solves the dual problem of the enhancement of contours, as is suggested by experimental observations. Its behavior is compared with psychophysical and physiological data on segmentation, contour enhancement, and contextual influences. We discuss the implications of segmentation without classification and the predictions of our V1 model, and relate it to other phenomena such as asymmetry in visual search.

In this thesis, the image segmentation methods based on the Mumford&amp / #8211 / Shah variational approach have been studied. By obtaining an optimum point of the Mumford-Shah functional which is a piecewise smooth approximate image and a set of edge curves, an image can be decomposed into regions. This piecewise smooth approximate image is smooth inside of regions, but it is allowed to be discontinuous region wise. Unfortunately, because of the irregularity of the Mumford Shah functional, it cannot be directly used for image segmentation. On the other hand, there are several approaches to approximate the Mumford-Shah functional. In the first approach, suggested by Ambrosio-Tortorelli, it is regularized in a special way. The regularized functional (Ambrosio-Tortorelli functional) is supposed to be gamma-convergent to the Mumford-Shah functional. In the second approach, the Mumford-Shah functional is minimized in two steps. In the first minimization step, the edge set is held constant and the resultant functional is minimized. The second minimization step is about updating the edge set by using level set methods. The second approximation to the Mumford-Shah functional is known as the Chan-Vese method. In both approaches, resultant PDE equations (Euler-Lagrange equations of associated functionals) are solved by finite difference methods. In this study, both approaches are implemented in a MATLAB environment. The overall performance of the algorithms has been investigated based on computer simulations over a series of images from simple to complicated.

Made available in DSpace on 2016-06-02T19:05:38Z (GMT). No. of bitstreams: 1 2578.pdf: 4936351 bytes, checksum: c3169360ab91e26c8efef7226a907cdf (MD5) Previous issue date: 2009-06-05 / Universidade Federal de Sao Carlos / In Brazil, the development of tools for weeds recognition, capable of aiding risk detection and decision making on the fieldwork is still embryonic. This master s thesis presents the development of a pattern recognition system that recognizes weeds and gives the occupation percentage of wide and narrow leaves in an agricultural production system, with digital image processing techniques. The development was based on considerations about image acquisition, pre-processing, texture based segmentation, descriptors for weeds recognition and occupation percentage of each kind of leaf. The validation has been developed considering geometric patterns generated in laboratory, as well as others obtained of a maize (Zea mays) production agricultural environment, i. e. two species of weeds, one with wide leaves (Euphorbia heterophylla L.) and other with narrow leaves (Digitaria sanguinalis Scop.). The results show recognition of about 84.24 percent for wide leaves and 80.17 percent for narrow leaves in agricultural environment and also the capability to spot weed on unreachable locations by natural vision. Besides, the method presents application in precision agriculture to improve the decision making in pulverization processes. / No Brasil é ainda embrionário o desenvolvimento de ferramentas de reconhecimento de plantas invasoras, capazes de auxiliar a tomada de decisão e indicar o seu risco no sistema de produção. Este trabalho apresenta o desenvolvimento de um sistema de reconhecimento de padrões de plantas invasoras e percentuais de ocupação de folhas largas e folhas estreitas, em sistemas de produção agrícola, utilizando técnicas de processamento digital de imagens. Para o desenvolvimento houve a consideração das etapas de aquisição das imagens, pré-processamento, segmentação baseada em textura, descritores para o reconhecimento das plantas invasoras e percentual de ocupação de cada tipo de planta. A validação foi desenvolvida considerando padrões geométricos gerados em laboratório, bem como o próprio ambiente de produção agrícola de milho (Zea mays), tomando por base duas espécies de plantas invasoras, sendo uma de folha larga (Euphorbia heterophylla L.), e outra de folha estreita (Digitaria sanguinalis Scop.). Resultados indicam uma taxa de acerto no reconhecimento em ambiente de campo da ordem de 84,24% para folhas largas e da ordem de 80,17% para folhas estreitas, além da capacidade de identificar plantas invasoras em locais restritos a visão natural. Adicionalmente, o resultado obtido apresenta potencial para a aplicação no manejo baseado em agricultura de precisão, o que auxilia na tomada de decisão em pulverização agrícola.

Made available in DSpace on 2014-12-17T15:48:04Z (GMT). No. of bitstreams: 1 TiagoSS_DISSERT.pdf: 2900373 bytes, checksum: ea7bd73351348f5c75a5bf4f337c599f (MD5) Previous issue date: 2012-08-17 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / The segmentation of an image aims to subdivide it into constituent regions or objects that have some relevant semantic content. This subdivision can also be applied to videos. However, in these cases, the objects appear in various frames that compose the videos. The task of segmenting an image becomes more complex when they are composed of objects that are defined by textural features, where the color information alone is not a good descriptor of the image. Fuzzy Segmentation is a region-growing segmentation algorithm that uses affinity functions in order to assign to each element in an image a grade of membership for each object (between 0 and 1). This work presents a modification of the Fuzzy Segmentation algorithm, for the purpose of improving the temporal and spatial complexity. The algorithm was adapted to segmenting color videos, treating them as 3D volume. In order to perform segmentation in videos, conventional color model or a hybrid model obtained by a method for choosing the best channels were used. The Fuzzy Segmentation algorithm was also applied to texture segmentation by using adaptive affinity functions defined for each object texture. Two types of affinity functions were used, one defined using the normal (or Gaussian) probability distribution and the other using the Skew Divergence. This latter, a Kullback-Leibler Divergence variation, is a measure of the difference between two probability distributions. Finally, the algorithm was tested in somes videos and also in texture mosaic images composed by images of the Brodatz album / A segmenta??o de uma imagem tem como objetivo subdividi-la em partes ou objetos constituintes que tenham algum conte?do sem?ntico relevante. Esta subdivis?o pode tamb?m ser aplicada a um v?deo, por?m, neste, os objetos est?o presentes nos diversos quadros que comp?em o v?deo. A tarefa de segmentar uma imagem torna-se mais complexa quando estas s?o compostas por objetos que contenham caracter?sticas texturais, com pouca ou nenhuma informa??o de cor. A segmenta??o difusa, do Ingl?s fuzzy, ? uma t?cnica de segmenta??o por crescimento de regi?es que determina para cada elemento da imagem um grau de pertin?ncia (entre zero e um) indicando a confian?a de que esse elemento perten?a a um determinado objeto ou regi?o existente na imagem, fazendo-se uso de fun??es de afinidade para obter esses valores de pertin?ncia. Neste trabalho ? apresentada uma modifica??o do algoritmo de segmenta??o fuzzy proposto por Carvalho [Carvalho et al. 2005], a fim de se obter melhorias na complexidade temporal e espacial. O algoritmo foi adaptado para segmentar v?deos coloridos tratando-os como volumes 3D. Para segmentar os v?deos, foram utilizadas informa??es provenientes de um modelo de cor convencional ou de um modelo h?brido obtido atrav?s de uma metodologia para a escolha dos melhores canais para realizar a segmenta??o. O algoritmo de segmenta??o fuzzy foi aplicado tamb?m na segmenta??o de texturas, fazendo-se uso de fun??es de afinidades adaptativas ?s texturas de cada objeto. Dois tipos de fun??es de afinidades foram utilizadas, uma utilizando a distribui??o normal de probabilidade, ou Gaussiana, e outra utilizando a diverg?ncia Skew. Esta ?ltima, uma varia??o da diverg?ncia de Kullback- Leibler, ? uma medida da diverg?ncia entre duas distribui??es de probabilidades. Por fim, o algoritmo foi testado com alguns v?deos e tamb?m com imagens de mosaicos de texturas criadas a partir do ?lbum de Brodatz e outros