We investigate the problem of learning optimal descriptors for a given classification task. Many hand-crafted descriptors have been proposed in the literature for measuring visual similarity. Looking past initial differences, what really distinguishes one descriptor from another is the tradeoff that it achieves between discriminative power and invariance. Since this trade-off must vary from task to task, no single descriptor can be optimal in all situations. Our focus, in this paper, is on learning the optimal tradeoff for classification given a particular training set and prior constraints. The problem is posed in the kernel learning framework. We learn the optimal, domain-specific kernel as a combination of base kernels corresponding to base features which achieve different levels of trade-off (such as no invariance, rotation invariance, scale invariance, affine invariance, etc.) This leads to a convex optimisation problem with a unique global optimum which can be solved for efficiently. The method is shown to achieve state-of-the-art performance on the UIUC textures, Oxford flowers and Caltech 101 datasets. 1.

We conjecture that texture can be characterized by the statistics of the wavelet detail coefficients and therefore introduce 2 feature sets: 1) the wavelet histogram signatures which capture all first order statistics using a model based approach; 2) the wavelet cooccurrence signatures, which reflect the coefficients' second order statistics. The introduced feature sets outperform the traditionally used energy. Best performance is achieved by combining histogram and cooccurrence signatures.

We propose an image segmentation method based on texture analysis. Our method is composed of two parts. The first part determines a novel set of texture features derived from a Gaussian--Markov random fields (GMRF) model. Unlike a GMRFbased approach, our method does not employ model parameters as features or require the extraction of features for a fixed set of texture types a priori. The second part is a two-dimensional (2--D) array of locally excitatory globally inhibitory oscillator networks (LEGION). After being filtered for noise suppression, features are used to determine the local couplings in the network. When LEGION runs, the oscillators corresponding to the same texture tend to synchronize, whereas different texture regions tend to correspond to distinct phases. In simulations, a large system of differential equations is solved for the first time using a recently proposed method for integrating relaxation oscillator networks. We provide results on real texture images to demonstrate the performance of our method. Index Terms---Dynamical systems, Gaussian Markov random fields, LEGION, neural networks, relaxation oscillators, texture segmentation. I.

This paper presents an autonomous agent-based image segmentation approach. In this approach, a digital image is viewed as a two-dimensional cellular environment in which the agents inhabit and attempt to label homogeneous segments. In so doing, the agents rely on some reactive behaviors such as breeding and diffusion. The agents that are successful in finding the pixels of a specific homogeneous segment will breed offspring agents inside their neighboring regions. Hence, the offspring agents will become likely to find more homogeneous-segment pixels. In the mean time, the unsuccessful agents will be inactivated, without further search in the environment. Index terms: Distributed autonomous agents, reactive behavior, evolutionary computation, breeding, diffusion, homogeneous-segment searching, adaptive image segmentation, and agent dynamics. I. INTRODUCTION The work to be presented in this paper explores an autonomous agent-based approach to image segmentation. In image segmentation, o...

| Based on a local spatial/frequency representation, we propose a spectral histogram as a feature statistic for characterizing texture appearance. The spectral histogram consists of marginal distributions of responses of a bank of lters and encodes implicitly the structure of images. The distance between two spectral histograms is measured using 2 -statistic. The spectral histogram with the associated distance measure exhibits several properties that are necessary for texture discrimination and classication. The spectral histogram provides a generic feature for texture as well as non-texture images, where the uniform image is a special case with a unique pattern. The spectral histogram is a nonlinear operator, consistent with the nonlinearity in human perception. Our classication experiments reveal that it can generalize well even with a small number of training samples and the classication result does not depend on a particular form of distance measure. We have obtained very g...

This paper proposes neural network-based text locations in complex color images. Texture information extracted on several color bands using neural networks is combined and corresponding text location algorithms are then developed. Text extraction filters can be automatically constructed using neural networks. Comparisons with other text location methods are presented; indicating that the proposed system has a better accuracy. <3 2001 Elsevier Science B.V. All rights reserved.

In this paper, we investigate material classification from single images obtained under unknown viewpoint and illumination. It is demonstrated that materials can be classified using the joint distribution of intensity values over extremely compact neighbourhoods (starting from as small as 3×3 pixels square), and that this outperforms classification using filter banks with large support. It is also shown that the performance of filter banks is inferior to that of image patches with equivalent neighbourhoods. We develop novel texton based representations which are suited to modelling this joint neighbour-hood distribution for MRFs. The representations are learnt from training images, and then used to classify novel images (with unknown viewpoint and lighting) into texture classes. Three such representations are proposed, and their performance is assessed and compared to that of filter banks. The power of the method is demonstrated by classifying 2806 images of all 61 materials present in the Columbia-Utrecht database. The classification performance surpasses that of recent state of the art filter bank based classifiers such as Leung and Malik (IJCV 01), Cula and Dana (IJCV 04), and Varma and Zisserman (IJCV 05). We also benchmark performance by classifying all the textures present in the Microsoft Textile database as well as the San Francisco outdoor dataset. We conclude with discussions on why features based on compact neighbourhoods can correctly discriminate between textures with large global structure and why the performance of filter banks is not superior to the source image patches from which they were derived.

Abstract. Methods for digital-image texture analysis are reviewed based on available literature and research work either carried out or supervised by the authors. The review has been prepared on request of Dr Richard Lerski, Chairman of the Management Committee of the COST B11 action “Quantitation of Magnetic Resonance Image Texture”.

Computer vision algorithms are natural candidates for high performance computing due to their inherent parallelism and intense computational demands. For example, a simple 3 x 3 convolution on a 512 x 512 gray scale image at 30 frames per second requires 67.5 million multiplications and 60 million additions to be performed in one second. Computer vision tasks can be classified into three categories based on their computational complexity andcommunication complexity: low-level, intermediate-level and high-level. Special-purpose hardware provides better performance compared to a general-purpose hardware for all the three levels of vision tasks. With recent advances in very large scale integration (VLSI) technology, an application specific integrated circuit (ASIC) can provide the best performance in terms of total execution time. However, long design cycle time, high development cost and inflexibility of a dedicated hardware deter design of ASICs. In contrast, field programmable gate arrays (FPGAs) support lower design verification time and easier design adaptability atalower cost. Hence, FPGAs with an array of reconfigurable logic blocks canbevery useful compute elements. FPGA-based custom computing machines are

We present a novel framework for unsupervised texture segmentation, which relies on statistical tests as a measure of homogeneity. Texture segmentation is formulated as a pairwise data clustering problem with a sparse neighborhood structure. The pairwise dissimilarities of texture blocks are computed using a multiscale image representation based on Gabor filters, which are tuned to spatial frequencies at different scales and orientations. We derive and discuss a family of objective functions to pose the segmentation problem in a precise mathematical formulation. An efficient optimization method, known as deterministic annealing, is applied to solve the associated optimization problem. The general framework of deterministic annealing and meanfield approximation is introduced and the canonical way to derive efficient algorithms within this framework is described in detail. Moreover the combinatorial optimization problem is examined from the viewpoint of scale space theory. The novel algorithm has been extensively tested on Brodatz-like microtexture mixtures and on real--word images. In addition, benchmark studies with alternative segmentation techniques are reported.