Illumination invariants based on Markov random fields

Content-based image retrieval (CBIR) systems, target database images using feature similarities with respect to the query. We introduce fast and robust image retrieval measures that utilise novel illumination invariant features extracted from three different Markov random field (MRF) based texture representations. These measures allow retrieving images with similar scenes comprising colour textured objects viewed with different illumination brightness or spectrum.The proposed illumination insensitive measures are compared favourably with the most frequently used features like the Local Binary Patterns, steerable pyramid and Gabor textural features, respectively. The superiority of these new illumination invariant measures and their robustness to added noise are empirically demonstrated in the illumination invariant recognition of textures from the Outex database.

An exceptional 3D wide-sense Markov model which can be completely solved analytically and easily synthesized is presented. The model can be modified to faithfully represent complex local data by adaptive numerically robust recursive estimators of all its statistics. Illumination invariants can be derived from some of its recursive statistics and exploited in content based image retrieval, supervised or unsupervised image recognition. Its modeling efficiency is demonstrated on several analytical and modeling image applications, in particular on unsupervised image or range data segmentation, bidirectional texture function (BTF) synthesis and compression, dynamic texture synthesis and adaptive multispectral and multichannel image and video restoration.

The analysis of image feature robustness using cometcloud

The highest fidelity representations of realistic real-world materials currently used comprise Bidirectional Texture Functions (BTF). The BTF is a six dimensional function depending on view and illumination directions as well as on planar texture coordinates. The huge size of such measurements, typically in the form of thousands of images covering all possible combinations of illumination and viewing angles, has prohibited their practical exploitation and obviously some compression and modelling method of these enormous BTF data spaces is inevitable. The proposed approach combines BTF spatial clustering with cluster index modelling by means of an efficient Markov random field model. This method allows to generate seamless cluster index of arbitrary size to cover large virtual 3D objects surfaces. The method represents original BTF data using a set of local spatially dependent Bidirectional Reflectance Distribution Function (BRDF) values which are combined according to synthesised cluster index and illumination / viewing directions. BTF data compression using this method is about 1:100 and their synthesis is very fast.KeywordsContextual NeighbourTexture SynthesisViewing AngleIllumination DirectionCluster RepresentativeThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

There have been much interest and a large amount of research on content based image retrieval (CBIR) in recent years due to the ever increasing number of digital images. Texture features play a key role in CBIR. Many texture features exist in literature, however, most of them are neither rotation invariant nor robust to scale and other variations. Texture features based on Gabor filters have been shown with significant advantages over other methods, and they are adopted by MPEG-7 as one of the texture descriptors for image retrieval. In this paper, we propose a rotation invariant curvelet features for texture representation. With systematic analysis and rigorous experiments, we show that the proposed curvelet texture features significantly outperforms the widely used Gabor texture features. A novel region padding method is also proposed to apply curvelet transform to region based image retrieval. Retrieval results from standard image databases show that curvelet features are promising for both texture and region representation.

In this chapter, a content-based image retrieval technique based on the concept of region-based image retrieval has been described. This technique integrates color, texture, and shape features using local binary patterns (LBPs). In this technique, the image is divided into a fixed number of blocks and from each block LBP-based color, texture, and shape features are computed. The color and texture features are extracted using LBP histograms of quantized color image and gray-level images, respectively. Shape features are computed using the binary edge map obtained using Sobel edge detector from each block. All three features are combined to make a single completed binary region descriptor (CBRD) represented in the LBP way. To support region-based retrieval, an effective region code-based scheme is employed. In this technique, the spatial relative locations of objects are also considered to increase the retrieval accuracy.

In this work, we investigate the visual appearance of real-world surfaces and the dependence of appearance on the geometry of imaging conditions. We discuss a new texture representation called the BTF (bidirectional texture function) which captures the variation in texture with illumination and viewing direction. We present a BTF database with image textures from over 60 different samples, each observed with over 200 different combinations of viewing and illumination directions. We describe the methods involved in collecting the database as well as the importqance and uniqueness of this database for computer graphics. A related quantity to the BTF is the familiar BRDF (bidirectional reflectance distribution function). The measurement methods involved in the BTF database are conducive to simultaneous measurement of the BRDF. Accordingly, we also present a BRDF database with reflectance measurements for over 60 different samples, each observed with over 200 different combinations of viewing and illumination directions. Both of these unique databases are publicly available and have important implications for computer graphics.

Content based image retrieval (CBIR) is an extrusive technique of retrieving the relevant images from vast image archives by extracting their low level features. In this research paper, the pursuance of five most prominent texture feature extraction techniques used in CBIR systems are experimentally compared in detail. The main issue with the CBIR systems is the proper selection of techniques for the extraction of low level features which comprises of color, texture and shape. Among these features, texture is one of the most decisive and dominant features. This selection of features completely depends upon the type of images to be retrieved from the database. The texture techniques explored here are Grey level co-occurrence matrix (GLCM), Discrete wavelet transform (DWT), Gabor transform, Curvelet and Local binary pattern (LBP). These are experimented on three touchstone databases which are Wang, Corel-5 K and Corel-10 K. The chief parameters of CBIR systems are evaluated here such as precision, recall and F-measure on all these databases using all the techniques. After detailed investigation it is figured out that LBP, GLCM and DWT provide highlighted and comparable results in all these datasets in terms of average precision. Besides practical implementation, the precised conceptual examination of these three texture techniques is also proposed in this article. So, this analysis is extremely beneficial for selecting the appropriate feature extraction technique by taking into consideration the experimental results along with image conditions such as noise, rotation etc.

Interpreting medical images is certainly a complex task which requires extensive knowledge [1]. According to Computer Aided Diagnosis (CAD) serves as a second opinion that will help radiologists in diagnosis and on the other hand Content-based Image Retrieval uses visual content to help users browse, search and retrieve similar medical images from a database based on the user’s interest [2-4]. The competency of the CBMIR system depends on feature extraction methods [5]. The textural features are very important to determine the content of a medical image. Textural features provide scenic depth, the spatial distribution of tonal variation, and surface orientation [6]. Therefore, this study seeks to compare and evaluate some of the hand-crafted texture feature extraction techniques in CBMIR. This is to help those concerned in enhancing CBIR systems to make informed decisions concerning the selection of the best textural feature extraction techniques. Since there is no clear indication of which of the various texture feature extraction techniques is best suited for a given performance metric when considering which of the techniques to choose for a particular study in CBMIR systems. The objective of this work, therefore, is to comparatively evaluate the performance of the following texture feature extraction techniques; Local Binary Pattern (LBP), Gabor Filter, Gray-Level Co-occurrence Matrix (GLCM), Haralick Descriptor, Features from Accelerated Segment Test (FAST) and Features from Accelerated Segment Test and Binary Robust Independent Elementary Features (FAST &BRIEF) using the metrics; precision, recall, f1-score, mean squared error (MSE), accuracy and time. These techniques are coupled with specific similarity measure to obtain results. The results showed that LBP, Haralick Descriptor, FAST, and GLCM had the best results in terms of (Precision and Accuracy), Time, F1-Score, and Recall respectively. LBP had 82.05% and 88.23% scores for precision and accuracy respectively. The following scores represent the performance of the Haralick descriptor, FAST, and GLCM models respectively; 0.88s, 38.7%, and 44.82%. These test scores are obtained from datasets ranging from 1k-10.5k. Aside from LBP outperforming the other 5 models mentioned, it still outperformed the following proposed models ' [7]’, ‘Tamura texture feature and wavelet transform combined with Hausdorff distance - [8]’, ‘ [9]’ in terms of (precision, accuracy, and recall) and (precision and recall) respectively and probably f1-score (since f1-score is the weighted average of precision and recall). It is believed that an ensemble of LBP, Haralick descriptors, and Support Vector Machine (SVM) can represent a robust system for both medical image retrieval and classification.

More and more attention has been paid to the invariant texture analysis, because the training and testing samples generally have not identical or similar orientations, or are not acquired from the same viewpoint in many practical applications, which often has negative influences on texture analysis. Local binary pattern (LBP) has been widely applied to texture classification due to its simplicity, efficiency, and rotation invariant property. In this paper, an integrated local binary pattern (ILBP) scheme including original rotation invariant LBP, improved contrast rotation invariant LBP, and direction rotation invariant LBP is proposed which can effectively overcome the deficiency of original LBP that is ignoring contrast and direction information. In addition, for surmounting another major drawback of LBP such as locality which can result in the lack of shape and space expression of the holistic texture image, Zernike moment features are fused into the improved LBP texture features in the proposed method because they comprise orthogonal and rotation invariant property and can be easily and rapidly calculated to an arbitrary high order. Experimental results show that the proposed method can be remarkably superior to the other state-of-the-art methods when rotation invariant texture features are extracted and classified.

Facial expression which carries rich information of body behavior is the leading carrier of human affective and the symbol of intelligence. The main purpose of this paper is to recognize 3D human facial expression. The research in this paper includes the expression feature extraction algorithm and fusion with different kinds of feature. To contain more local texture feature information, we proposed a new feature of 3D facial expression named Local Threshold Binary Pattern (LTBP) which based on Local Binary Pattern (LBP). We calculate the difference of gray value standard between neighboring pixels and the center pixel as a threshold to binary instead of the traditional LBP operation which only comparison of size between neighboring pixels and the center pixel. After we get the LTBP feature, we fuse the LTBP and HOG (Histogram of Oriented Gradient) features to get multi-feature fusion for 3D facial expression recognition. Our algorithm of 3D facial expression recognition comprises three steps: (1) extracting two sets of feature vectors and establishing the correlation criterion function between the two sets of feature vectors; (2) solving the two sets canonical projective vectors and extracting their canonical correlation features by the framework of canonical correlation analysis algorithm; (3) doing feature fusion for classification by using proposed strategy. We have performed comprehensive experiments on the BU-3DFE database which is presently the largest available 3D face database. We have achieved verification rates of more than 90% for the 3D facial expression recognition.

Histogram refinement for texture descriptor based image retrieval

In response to the current need for positive identification of cattle traceability, this paper presents a novel facial representation model of cattle based on local binary pattern (LBP) texture features and some extended LBP descriptors are also introduced. Algorithm training was performed independently on several normalized gray face images of 30 cattle (with each having a set of six, seven, eight, and nine images respectively). Robust alignment by sparse and low-rank decomposition was also used to align the images because of variations in illumination, image misalignment and occlusion in the test image. The performance of this technique was assessed on a separate set of images using the weighted Chi square distance [1]. The LBP descriptor shows its excellence in efficiency and accuracy with regard to the encouraging results on cattle face recognition. More training sets and modified algorithms will be considered to improve recognition rates. Future work should aim at improving the automation of the system and combining the LBP histogram with other effective histograms.

Brain diseases can cause invisible disorders, cognitive and behavioral changes. Their symptoms vary widely. In some cases, treatment can improve the symptoms while in other cases injuries become permanent. Many disorders are progressive. Therefore, early and accurate diagnosis of disorder is essential for improving disorder condition and patient’s quality of life. This paper presents the brain disease diagnosis system in which two feature extraction methods are compared. One of the feature extraction methods uses local binary pattern and steerable pyramid (SP) to decompose magnetic resonance (MR) brain images into subbands which are termed as LBPSP subbands. Another feature extraction method uses SP solely to decompose MR images into SP subbands. Energies over LBPSP and SP subbands are calculated. The features are subjected to backpropagation neural network classifier. To prove the effectiveness of the proposed system, multi-class disease classification is carried out on four MR image datasets. Also, ‘one-vs-all’ binary classification is performed on one of the datasets. Energy features of LBPSP subbands achieve multi-class classification accuracies of 97.67%, 97.27%, 94.67% and 85.01% on datasets DS-200, DS-310, DS-255 and DS-612, respectively. The performance measures of ‘one-vs-all’ binary class classification prove the competency and efficiency of LBPSP subband features over the existing methods.The comparative results of two feature extraction methods indicate that the energy features of LBPSP subbands have more discriminating potential than energy features of SP subbands. Experimental results reveal that energy features of LBPSP subbands lead to the existing classification methods.

ABSTRACTMultivariate image analysis was used to estimate the arsenic concentrations in froths resulting from the flotation of different mixtures of realgar and orpiment particles in a laboratory batch flotation cell. The realgar floated rapidly and in excess of 90% of the mineral could be recovered after 2 minutes, whereas only 48–75% of the orpiment could be recovered in the same time. Textural features, based on grey level co-occurrence matrices (GLCMs), local binary patterns (LBPs), steearable pyramids and textons were used in the analysis. Random forest models could explain approximately 71–77% of the variance in the arsenic using either of the texton, steerable pyramid or LBP features. This was considerably better than what could be obtained with the GLCM features. Monitoring of froth flotation cells was simulated with the batch data. The texton textural features were the most discriminatory with regard to detecting changes in the arsenic content of the froth.