Local Texture Analysis

Automatic identification of male and female from facial image allows many useful applications in biometrics, surveillance, and human-computer interaction. This paper presents a robust face feature descriptor for gender classification from facial image. The proposed method is based on the compound local binary pattern (CLBP), an extension of the LBP texture operator. The CLBP operator exploits 2P bits to encode the information of a local neighborhood of P neighbors, where P bits are used to express the sign information and the other P bits are used to express the magnitude information of the differences between the center and the neighbor gray values. The performance of the proposed method has been evaluated using a large dataset comprising 1800 facial images collected from the FERET database. Extensive experiments with support vector machine classifier show the superiority of the CLBP feature descriptor against some well-known texture operators.

Grey Level Co-occurrence Matrices (GLCM) are one of the earliest techniques used for image texture analysis. In this paper we defined a new feature called trace extracted from the GLCM and its implications in texture analysis are discussed in the context of Content Based Image Retrieval (CBIR). The theoretical extension of GLCM to n-dimensional gray scale images are also discussed. The results indicate that trace features outperform Haralick features when applied to CBIR.

Texture is one of the most important visual characteristics of image. Texture classification is a process of assigning unknown texture to known texture class. Applications areas of texture classification are medical image analysis, object recognition, biometrics, content based image retrieval, remote sensing, industrial inspection, document analysis and many more. In this paper we discussed some feature extraction and classification methods used for texture classification namely, Local binary pattern, Scale invariant feature transform, Speed up robust feature, Fourier transformation, Texture spectrum, Gray level co-occurrence matrix, K-nearest neighbor, Artificial neural network and Support Vector Machine. We also discussed some popular texture datasets Brodatz, Outex, CUReT and VisTex used for texture classification.

The article considers the specifics of functioning of a Roman text in the book of Peter Vail's "Genius of Place" (1999), referred to basic semantic constants of Roman text presented in the book and the associated motive-shaped complex, revealed their artistic originality of Vail’s Rimlyane.

The article deals with the specifics of the functioning of the city text in the book of Peter Vail "Genius of Place" (1999); reveals a motivic-shaped complex of Spanish urban texts; defines their basic semantic constants, due, on the one hand, ethnographic stereotypes and genre-style traditions of travel prose, on the other hand, – philosophical and aesthetic settings of the author; reconstructs the individual-author's version of the unified image of Spain.

In this article the representation of Parisian text in The Letters of Russian traveller by N.M.Karamzin and The Genius of Place by P.Vail is considered. The comparison of two travel allows to speak about the unity and continuity of this supertext. Particular attention is paid to the differences in the worldview of both writers, to determine the difference in the perception of Paris and its artistic expression.

"The aim of this paper is to present a new approach for local
texture analysis called LFP (Local Fuzzy Pattern). The
proposed method employs fuzzy numbers to deal with the
uncertainties from gray levels of image pixels, and calculates
the membership degree of pixels in relation to a texture
region. The performance of our approach has been compared
to the local texture descriptor called LBP (Local Binary
Pattern) showing an error rate of 2.5% against 15% for the
LBP considering the same sample size from the Brodatz album."

Grey Level Co-occurrence Matrices (GLCM) are one of the earliest techniques used for image texture analysis. In this paper we defined a new feature called trace extracted from the GLCM and its implications in texture analysis are discussed in the context of Content Based Image Retrieval (CBIR). The theoretical extension of GLCM to n-dimensional gray scale images are also discussed. The results indicate that trace features outperform Haralick features when applied to CBIR.

Представление о «родной земле» -как и вообще о «земле» -мало ассоциируется. с городом. Тем не менее в число задач патриотического воспитания в советскую эпоху входило и развитие городского локального патриотизма. Одним из способов выполнения этой задачи была популяризация особого жанра, который можно обозначить как «песня о городе». Количество написанных за по лвека ; развитого социализма и последующие полтора . десятилетия песен о городах приближается к тысяче (заметим в скобках, что в наши дни эта традиция, почти не видоизменившаяся с советских времен, получила второе дыхание). Сочинение городских песен всячески приветствовалось, поощрялось и стимулировалось «сверху». Так, в 1950-х -1970-х годах издавались многочисленные сборники, составленные по тематическому принципу из песен о том или ином городе: «Песни об Одессе», «Легендарный Севастополь», «Тула -родина моя» и т. п. Кроме того, выпускались сборники песен местных композиторов (например одесских, томских, и проч.), составлявшиеся по результатам специально организованных конкурсов. Налицо элемент государственной политики, направленной, как можно предположить, на то, чтобы каждый советский человек знал и уважал как свой город, так и другие города своей большой страны.

The aim of this paper is to present a new approach for local texture analysis called LFP (Local Fuzzy Pattern). The proposed method employs fuzzy numbers to deal with the uncertainties from gray levels of image pixels, and calculates the membership degree of pixels in relation to a texture region. The performance of our approach has been compared to the local texture descriptor called LBP (Local Binary Pattern) showing an error rate of 2.5% against 15% for the LBP considering the same sample size from the Brodatz album.

Deriving an effective facial feature from original face images is a vital step for a successful automatic facial expression recognition. In this paper, we proposed a new feature descriptor, Local Principal Texture Pattern (LPTP), for expression recognition. We compute the LPTP feature, at each pixel, by extracting the principal directions of the local neighborhood, and we code the intensity differences on these directions. The mixture of direction and contrast information makes our descriptor robust against rotation and illumination changes.  Consequently, we represent each expression as a distribution of LPTP codes. Our experiments demonstrate the superiority of the proposed feature, on two facial expression databases, over the existing methods.

In this paper, we will focus on the Spatial Gray Level Dependence Matrices SGLDM to extract the Haralick's texture features of the ultrasound breast lesions. This method relies on the manual selection of the region of interest, which results in the dependence of parameters values on the extracted region. For that reason, an improved Spatial Gray Level Dependence Matrices based on the segmented masses using active contour was applied. This method outperforms the existing SGLDM method because it allows establishing a well determined threshold for the classification of lesions.

The study gives an overview on basic types of crystallographic textures developing during hot rolling of polycrystalline steels. The results are grouped into three fundamental classes of textures.
The first group comprises pure Fe, some weakly bonded B2  and BO3 structured intermetallics, as well as closely related alloys such as ferritic low carbon and microalloyed interstitial free steels. The second group Includes highly alloyed corrosion-resistant ferritic stainless and Fe-Si transformer steels. Typical examples are steels with about 10 -17 wl.% Cr. with about 3 wt.% Si, as well as body centered cubic transition metals such as Ta. Mo, and Nb which do not undergo any phase transformation during hot rolling. The third group comprises stable and instable austenitic stainless steels for instance on the basis of larger amounts of Cr and Ni or on Mn as well as duplex steels. Most L 12 structured intermetallic alloys can also be assigned to this group. The suggested classification scheme is discussed in terms of different processing parameters, thermodynamics. microstructure, and crystallographic
aspects.

The conventional quantitative evaluation of dark lock-in thermography (DLIT), electroluminescence (EL), and photoluminescence (PL) images of solar cells is based on the model of independent diodes, where each image pixel is assumed to be connected with the terminals by an independent series resistance. In reality, however, the solar cell represents a 2-dimensional resistance-diode network. In this work solar cells containing well-defined spatial distributions of the saturation current density J 01 and also containing J 02 -type and ohmic shunts are modeled for various externally applied biases and illumination conditions realistically as a 2-dimensional resistance-diode network. The resulting local diode voltage distributions are converted into DLIT, EL and PL images, which are further processed by conventional evaluation methods, which rely on the simple model of independent diodes. These are the so-called "Local-IV" method for the DLIT analysis, which may be supported by EL results to obtain series resistance images, and "C-DCR" for the PL analysis. This leads to calculated images of the local effective series resistance R s and of J 01 . Regarding the resulting R s images, PL shows the expected series resistance distribution and is not affected by the shunt regions. The DLIT-EL R s images instead yield expected values only in the homogeneous regions, which are not affected by the assumed shunts. DLIT-EL determines higher values of R s in local shunt regions and lower values around these regions and in spatially extended shunt regions. Regarding the J 01 images both methods again give the expected results if J 01 is distributed homogeneously. However, in the shunted regions, PL suffers from balancing currents within the emitter and DLIT from optical blurring. By comparing local and extended regions of increased J 01 we find that DLIT approximates the expected J 01 value better than PL, which clearly underestimates even extended local maxima of J 01 . For a local current analysis of silicon solar cells we recommend the use of DLIT for the determination of J 01 images and PL for the determination of R s images.