Color Texture Descriptors Research Articles

Evaluating color texture descriptors under large variations of controlled lighting conditions.

The recognition of color texture under varying lighting conditions remains an open issue. Several features have been proposed for this purpose, ranging from traditional statistical descriptors to features extracted with neural networks. Still, it is not completely clear under what circumstances a feature performs better than others. In this paper, we report an extensive comparison of old and new texture features, with and without a color normalization step, with a particular focus on how these features are affected by small and large variations in the lighting conditions. The evaluation is performed on a new texture database, which includes 68 samples of raw food acquired under 46 conditions that present single and combined variations of light color, direction, and intensity. The database allows us to systematically investigate the robustness of texture descriptors across large variations of imaging conditions.

Influence of Texture and Colour in Breast TMA Classification.

Breast cancer diagnosis is still done by observation of biopsies under the microscope. The development of automated methods for breast TMA classification would reduce diagnostic time. This paper is a step towards the solution for this problem and shows a complete study of breast TMA classification based on colour models and texture descriptors. The TMA images were divided into four classes: i) benign stromal tissue with cellularity, ii) adipose tissue, iii) benign and benign anomalous structures, and iv) ductal and lobular carcinomas. A relevant set of features was obtained on eight different colour models from first and second order Haralick statistical descriptors obtained from the intensity image, Fourier, Wavelets, Multiresolution Gabor, M-LBP and textons descriptors. Furthermore, four types of classification experiments were performed using six different classifiers: (1) classification per colour model individually, (2) classification by combination of colour models, (3) classification by combination of colour models and descriptors, and (4) classification by combination of colour models and descriptors with a previous feature set reduction. The best result shows an average of 99.05% accuracy and 98.34% positive predictive value. These results have been obtained by means of a bagging tree classifier with combination of six colour models and the use of 1719 non-correlated (correlation threshold of 97%) textural features based on Statistical, M-LBP, Gabor and Spatial textons descriptors.

Integral split-and-merge methodology for real-time image segmentation

The segmentation of images is a critical step in many computer vision applications. Additionally, some applications require the achievement of acceptable segmentation quality while the algorithm is executed in real time. In this study, we present a split-and-merge segmentation methodology that uses integral images to improve the execution time. We call our methodology integral split and merge (ISM) segmentation. The integral images are used here to calculate statistics of the image regions in constant time. Those statistics are used to guide the splitting process by identifying the homogeneous regions in the image. We also propose a merge criterion that performs connected component analysis of the homogeneous regions. Moreover, the merging procedure is able to group regions of the image showing gradients. Furthermore, the number of regions resulting from the seg- mentation process is determined automatically. In a series of tests, we compare ISM against other state-of-the- art algorithms. The results from the tests show that our ISM methodology obtains image segmentations with a comparable quality, using a simple texture descriptor instead of a combination of color-texture descriptors. The proposed ISM methodology also has a piecewise linear computational complexity, resulting in an algorithm fast enough to be executed in real time. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, includ- ing its DOI. (DOI: 10.1117/1.JEI.24.1.013007)

Combining local binary patterns and local color contrast for texture classification under varying illumination

This paper presents a texture descriptor for color texture classification specially designed to be robust against changes in the illumination conditions. The descriptor combines a histogram of local binary patterns (LBPs) with a novel feature measuring the distribution of local color contrast. The proposed descriptor is invariant with respect to rotations and translations of the image plane and with respect to several transformations in the color space. We evaluated the proposed descriptor on the Outex test suite, by measuring the classification accuracy in the case in which training and test images have been acquired under different illuminants. The results obtained show that our descriptor outperforms the original LBP approach and its color variants, even when these are computed after color normalization. Moreover, it also outperforms several other color texture descriptors in the state of the art.

Quaternion Affine Moment Invariants for Color Texture Recognition

We introduce a generalization of the affine moment invariants for color textures recognition using Quaternion Algebra and present results of experiments on a challenging dataset. The advantage of the proposed method is that it can process color image directly, without losing color information. It is shown that the QAMIs can be obtained from the usual AMIs. Experiments show that color texture descriptors are more efficient and discriminant. Moreover, a new set of invariants can be obtained from each QAMI (due to the non-commutativity of Quaternion algebra).

Colour texture classification from colour filter array images using various colour spaces

This study focuses on the classification of colour textures acquired by single-sensor colour cameras. In such cameras, the colour filter array (CFA) makes each photosensor sensitive to only one colour component, and CFA images must be demosaiced to estimate the final colour images. We show that demosaicing is detrimental to the textural information because it affects colour texture descriptors such as chromatic co-occurrence matrices (CCMs). However, it remains desirable to take advantage of the chromatic information for colour texture classification. This information is incompletely defined in CFA images, in which each pixel is associated to a single colour component. It is hence a challenge to extract standard colour texture descriptors from CFA images without demosaicing. We propose to form a pair of quarter-size colour images directly from CFA images without any estimation, then to compute the CCMs of these quarter-size images. This allows us to compare textures by means of their CCM-based similarity in texture classification or retrieval schemes, with still the ability to use different colour spaces. Experimental results achieved on benchmark colour texture databases show the effectiveness of the proposed approach for texture classification, and a complexity study highlights its computational efficiency.

Multilayer graph cuts based unsupervised color–texture image segmentation using multivariate mixed student's t-distribution and regional credibility merging

This paper proposes an unsupervised color–texture image segmentation method. In order to enhance the effects of segmentation, a new color–texture descriptor is designed by integrating the compact multi-scale structure tensor (MSST), total variation (TV) flow, and the color information. Due to the fact that MSST does not work well for separating regions with large-scale texture, the total variation flow is used to auxiliarily describe the texture feature by extracting local scale information. To segment the color–texture image in an unsupervised and multi-label way, the multivariate mixed student's t-distribution (MMST) is chosen for probability distribution modeling, as MMST can describe the distribution of color–texture features accurately. Since the valid class number is hard to adaptively determine in advance, a component-wise expectation–maximization for MMST (CEM3ST) algorithm is proposed, which can effectively initialize the valid class number. Then, we can build up the energy functional according to the valid class number, and optimize it by multilayer graph cuts method. However, the problem of over/error-segmentation often happens. To overcome this problem, a strategy of regional credibility merging (RCM) is presented by integrating the regional adjacency relationship, region size, common edge between regions, and regional color–texture dissimilarity. In order to terminate the whole segmentation process, an adaptive iteration convergence criterion is designed, which combines the negative logarithm of probability of all color–texture features with the Kullback–Leibler (KL) divergence for MMST. Experiments using a large number of synthesis color–texture images and real natural scene images demonstrate the superiority of our proposed method, such as the effective over/error-segmentation reduction, high segmentation accuracy, and outperforming visual entirety/consistency.

LS-SVM-based image segmentation using pixel color-texture descriptors

Image segmentation remains an important, but hard-to-solve, problem since it appears to be application dependent with usually no a priori information available regarding the image structure. Moreover, the increasing demands of image analysis tasks in terms of segmentation results' quality introduce the necessity of employing multiple cues for improving image-segmentation results. In this paper, we present a least squares support vector machine (LS-SVM) based image segmentation using pixel color-texture descriptors, in which multiple cues such as edge saliency, color saliency, local maximum energy, and multiresolution texture gradient are incorporated. Firstly, the pixel-level edge saliency and color saliency are extracted based on the spatial relations between neighboring pixels in HSV color space. Secondly, the image pixel's texture features, local maximum energy and multiresolution texture gradient, are represented via nonsubsampled contourlet transform. Then, both the pixel-level edge color saliency and texture features are used as input of LS-SVM model (classifier), and the LS-SVM model (classifier) is trained by selecting the training samples with Arimoto entropy thresholding. Finally, the color image is segmented with the trained LS-SVM model (classifier). This image segmentation not only can fully take advantage of the human visual attention and local texture content of color image, but also the generalization ability of LS-SVM classifier. Experimental results show that our proposed method has very promising segmentation performance compared with the state-of-the-art segmentation approaches recently proposed in the literature.

Color Image Segmentation using Fuzzy Local Texture Patterns

Texture is one of the fundamental image characteristics useful in computer vision tasks such as object recognition and scene analysis. Texture segmentation is one of the image analysis tasks. The prospect of texture segmentation depends on the choice of the texture description method and the segmentation procedure. In this paper, color-texture descriptors are proposed to represent the texture contents of the color images. In these texture description schemes, small areas of the image are represented by fuzzy based local texture patterns and the entire image is represented by frequency occurrence of such texture patterns. Supervised segmentation of color images is performed using these color-texture descriptors and promising results are obtained. General Terms Computer Vision, Texture Analysis.

Low-dimensional and comprehensive color texture description

Image retrieval can be dealt by combining standard descriptors, such as those of MPEG-7, which are defined independently for each visual cue (e.g. SCD or CLD for Color, HTD for texture or EHD for edges). A common problem is to combine similarities coming from descriptors representing different concepts in different spaces. In this paper we propose a color texture description that bypasses this problem from its inherent definition. It is based on a low dimensional space with 6 perceptual axes. Texture is described in a 3D space derived from a direct implementation of the original Julesz’s Texton theory and color is described in a 3D perceptual space. This early fusion through the blob concept in these two bounded spaces avoids the problem and allows us to derive a sparse color-texture descriptor that achieves similar performance compared to MPEG-7 in image retrieval. Moreover, our descriptor presents comprehensive qualities since it can also be applied either in segmentation or browsing: (a) a dense image representation is defined from the descriptor showing a reasonable performance in locating texture patterns included in complex images; and (b) a vocabulary of basic terms is derived to build an intermediate level descriptor in natural language improving browsing by bridging semantic gap.

Image segmentation based on the integration of colour–texture descriptors—A review

The adaptive integration of the colour and texture attributes in the development of complex image descriptors is one of the most investigated topics of research in computer vision. The substantial interest shown by the research community in colour–texture-based segmentation is mainly motivated by two factors. The first is related to the observation that the imaged objects are often described at perceptual level by distinctive colour and texture characteristics, while the second is motivated by the large spectrum of possible applications that can be addressed by the colour–texture integration in the segmentation process. Over the past three decades a substantial number of techniques in the field of colour–texture segmentation have been reported and it is the aim of this article to thoroughly evaluate and categorise the most relevant algorithms with respect to the modality behind the integration of these two fundamental image attributes. In this paper we also provide a detailed discussion about data collections, evaluation metrics and we review the performance attained by state of the art implementations. We conclude with a discussion that samples our views on the field of colour–texture image segmentation and this is complemented with an examination of the potential future directions of research.

Image retrieval based on multi-texton histogram

This paper presents a novel image feature representation method, called multi-texton histogram (MTH), for image retrieval. MTH integrates the advantages of co-occurrence matrix and histogram by representing the attribute of co-occurrence matrix using histogram. It can be considered as a generalized visual attribute descriptor but without any image segmentation or model training. The proposed MTH method is based on Julesz's textons theory, and it works directly on natural images as a shape descriptor. Meanwhile, it can be used as a color texture descriptor and leads to good performance. The proposed MTH method is extensively tested on the Corel dataset with 15 000 natural images. The results demonstrate that it is much more efficient than representative image feature descriptors, such as the edge orientation auto-correlogram and the texton co-occurrence matrix. It has good discrimination power of color, texture and shape features.

Multivariate statistical projection methods to perform robust feature extraction and classification in surface grading

We present an innovative way to simultaneously perform feature extraction and classification for the quality-control issue of surface grading by applying two multivariate statistical projection methods: SIMCA and PLS-DA. These tools have been applied to compress the color texture data that describe the visual appearance of surfaces (soft color texture descriptors) and to directly perform classification using statistics and predictions from the projection models. Experiments have been carried out using an extensive ceramic images database (VxC TSG) comprised of 14 different models, 42 surface classes, and 960 pieces. A factorial experimental design evaluated all the combinations of several factors affecting the accuracy rate. These factors include the tile model, color representation scheme (CIE Lab, CIE Luv, and RGB), and compression/classification approach (SIMCA and PLS-DA). Moreover, a logistic regression model is fitted from the experiments to compute accuracy estimates and study the effect of the factors on the accuracy rate. Results show that PLS-DA performs better than SIMCA, achieving a mean accuracy rate of 98.95%. These results outperform those obtained in a previous work where the soft color texture descriptors in combination with the CIE Lab color space and the k-NN classifier achieved an accuracy rate of 97.36%.

Feature selection based-on genetic algorithm for image annotation

Machine learning techniques for feature selection, which include the optimization of feature descriptor weights and the selection of optimal feature descriptor subset, are desirable to enhance the performance of image annotation systems. In our system, the multimedia content description interface (MPEG-7) image feature descriptors consisting of color descriptors, texture descriptors and shape descriptors are employed to represent low-level image features. We use a real coded chromosome genetic algorithm and k-nearest neighbor ( k-NN) classification accuracy as fitness function to optimize the weights of MPEG-7 image feature descriptors. A binary one and k-NN classification accuracy combining with the size of feature descriptor subset as fitness function are used to select optimal MPEG-7 feature descriptor subset. Furthermore, a bi-coded chromosome genetic algorithm is used for the simultaneity of weight optimization and descriptor subset selection, whose fitness function is the same as that of the binary one. The experimental results over 2000 classified Corel images show that with the real coded genetic algorithm, the binary coded one and the bi-coded one, the accuracies of image annotation system are improved by 7%, 9% and 13.6%, respectively, comparing to the method without machine learning. Furthermore, 2 of 25 MPEG-7 feature descriptors are selected with the binary coded genetic algorithm and four with the bi-coded one, which may improve the efficiency of system significantly.

Performance evaluation of soft color texture descriptors for surface grading using experimental design and logistic regression

This paper presents a novel approach to the question of surface grading, the soft color texture descriptors method. This method is extracted from an extensive evaluation process of several factors based on the use of two well established statistical tools: experimental design and logistic regression. The utility of different combinations of factors is evaluated in regard to the problem of automatic classification of materials such as ceramic tiles that need to be grouped according to homogeneous visual appearance, that is, the surface grading application. The set of factors includes the number of neighbors in the k-NN classifier (several values of k parameter), color space representation schemes (CIE Lab, CIE Luv, RGB, and grayscale), and color texture features (mean, standard deviation, 2nd–5th histogram moments). A factorial experimental design is performed testing all combinations of the above factors on a large image database of ceramic tiles. Accuracy estimates are computed using logistic regression to determine the best combinations of factors. From the point of view of machine learning the overall process conforms a wrapper approach able to select significant design choices ( k parameter in k-NN classifier and color space) and carry out a feature selection within the set of color texture features at the same time. Experiments were repeated with alternate color texture schemes from the literature: color histograms and centile-LBP. Comparisons of methods are presented describing both accuracy estimates and runtimes.

COLOR IMAGE RETRIEVAL BASED ON COLOR-TEXTURE-EDGE FEATURE HISTOGRAMS

In this paper, two novel texture descriptors and two novel edge descriptors are proposed, which are low-dimension, effective, and are obtained by a relative simple approach. The two texture descriptors are the directional difference unit and the gradient unit histogram, which are rotation invariant. The two edge descriptors are the quantized max-min difference histogram and the quantized fuzzy entropy histogram, the former is more suitable for describing the images with relatively regular texture characteristic, the latter is more suitable for describing the images with relatively regular structure characteristic or no regular characteristic. Combing color descriptor, texture descriptor and edge descriptor to form hybrid visual feature index to retrieve natural color images, this method is insensitive to image rotation and translation. Experimental results show that the method achieves better performance than other recent relevant methods.

MPEG-7 Visual Descriptors—Contributions for Automated Feature Extraction in Capsule Endoscopy

Recent advances in miniaturization led to the development of what is now called the endoscopic capsule. This small device is swallowed by a patient and films the whole gastrointestinal tract, allowing the detection of abnormalities. Currently, a doctor typically needs up to two hours to analyze a full exam, so automation is desirable. This paper presents a methodology for measuring the potential of selected visual MPEG-7 descriptors for the task of specific medical event detection such as blood, ulcers. Experiments show that the best results are obtained by the Scalable Color and Homogenous Texture descriptors, especially if only relevant coefficients are used.