The research of computer aided mammography continues to be developed. For the mass lesions of breast,  presents a tool system in 2006, including imaging segmentation of ROI, extracting ROI characterization “by means of textural features computed from the gray tone spatialdependencematrix (GTSDM), containing second-order spatial statistics information on the pixel graylevel intensity”, and classify ROI with neural network. In 2008, Pal et al. used 24 kinds of features for four types of window sizes to detect microcalcification, which resulted in computing 87 features for each pixel. 
Various approaches have been proposed in the literature for texture characterization of images. Some of them are based on statistical properties, others on fractal measures and some more on multi-resolution analysis. Basically, these approaches have been applied on mono-band images. However, most of them have been extended by including the additional information between spectral bands to deal with multi-band texture images. In this article, we investigate the problem of texture characterization for multi-band images. Therefore, we aim to add spectral information to classical texture analysis methods that only treat gray-levelspatial variations. To achieve this goal, we propose a spatial and spectral grayleveldependence method (SSGLDM) in order to extend the concept of graylevel co-occurrence matrix (GLCM) by assuming the presence of texture joint information between spectral bands. Thus, we propose new multi-dimensional functions for estimating the second-order joint conditional probability density of spectral vectors. Theses functions can be represented in structure form which can help us to compute the occurrences while keeping the corresponding components of spectral vectors. In addition, new texture features measurements related to (SSGLDM) which define the multi-spectral image properties are proposed. Extensive experiments have been carried out on 624 textured multi-spectral images for use in prostate cancer diagnosis and quantitative results showed the efficiency of this method compared to the GLCM. The results indicate a significant improvement in terms of global accuracy rate. Thus, the proposed approach can provide clinically useful information for discriminating pathological tissue from healthy tissue.
A statistical method of examining texture that considers the spatial relationship of pixels is the gray- level co-occurrence matrix (GLCM), also known as the gray-levelspatialdependencematrix. The GLCM functions characterize the texture of an image by calculating how often pairs of pixel with specific values and in a specified spatial relationship occur in an image, creating a GLCM, and then extracting statistical measures from this matrix. The texture filter functions, described in Using Texture Filter Functions, cannot provide information about shape, i.e., the spatial relationships of pixels in an image. The gray-level co-occurrence matrix can reveal certain properties about the spatial distribution of the gray levels in the texture image. For example, if most of the entries in the GLCM are concentrated along the diagonal, the texture is coarse with respect to the specified offset.
d = 3, θ = 0°, 45°, 90° and 135° constant GLCM is calculated. So there are four co-occurrence matrices. According GLCM in each computer, the successful co-occurrence matrix, which characterizes the behavior of the statistical property is obtained 8. It features: Angular Second Moment, Contrast, Correlations, Dissimilarity, Entropy, Homogeneity, Maximum probability, Average. The equations of these features are below:
preprocessing techniques such as size normalization, skewing, noise removal, image enhancement and image segmentation. The preprocessing is the essential step for given input images. The output of first phase i.e. preprocessed image is send to next phase called feature extraction. In second phase the texture feature can be extracted using Gray-Level Co-Occurrence Matrix (GLCM) which is a popular statistical method of extracting textural feature from images. The four static parameters properties of Energy, Homogeneity, Correlation and Contrast are used to decrease the computational complexity. Texture features of nose and mouth extracted from given preprocessed image using GLCM. The static parameters of nose and mouth are calculated using GLCM. From the Nose part four important texture features and from Mouth part another four significant texture features are extracted and stored in a vector.
2) HSV Color Space: The HSV color space (Hue, Saturation, Value) is often used by people who are selecting colors (e.g., of paints or inks) from a color wheel or palette, because it corresponds better to how people experience color than the RGB color space does. As hue varies from 0 to 1.0, the corresponding colors vary from red through yellow, green, cyan, blue, magenta, and back to red, so that there are actually red values both at 0 and 1.0. As saturation varies from 0 to 1.0, the corresponding colors (hues) vary from unsaturated (shades of gray) to fully saturated (no white component). As value, or brightness, varies from 0 to 1.0, the corresponding colors become increasingly brighter. The following Fig. 3illustrates the HSV color space .
Content Based Image Retrieval (CBIR) approach which provides efficient and effective means to extract most similar images stored in the database based on image contents. It aims at retrieving images which are perceptually meaningful by making the best use of a single or combination of visual content descriptors. The major steps are feature extraction and similarity comparison. This paper expands along the two techniques “Haar and GLCM” on Texture based image retrieval. Texture analysis is one of the widely used operations for feature extraction. The first method explores texture features extraction of the image by applying Haar Wavelet Transform (HWT) and then applying Gradient Operator Concept on LH and HL bands of HWT. The implemented second method computes the texture features by applying popular graylevel co-occurrence matrix (GLCM) approach. The texture features thus extracted are subjected to similarity comparison and few images with higher similarity value are retrieved. The Brodatz image database that utilizes texture features forms the basis of the comparison and retrieval.It is concluded that the implemented GLCM provides much better results when compared with the popular Haar.
Abstract: Statistical features extracted from the GrayLevel Co-occurrence Matrix (GLCM) of liquid crystal textures are used to investigate the phase transition temperatures of nematic liquid crystals p – n Alkyl benzoic acids (nBA) where n = 8,9 and10. Textures of compounds are recorded as a function of temperature using Polarizing Optical Microscope attached to the hot stage and high resolution camera. In this method, second order statistical parameters – contrast, energy, homogeneity and correlation of the sample textures are computed using MATLAB software. The changes associated in the values of computed statistical features as a function of temperature is a helpful process to identify the phase transition temperatures of the samples. Results obtained from this method are compared with literature values of Differential Scanning Calorimetry (DSC) and are in agreement.
Magnetic Resonance Imaging (MRI) has been a robust tool for the diagnosis of brain tumors. MRI is an imaging technique that provides detailed information about brain anatomy. This paper announces a novel method for efficient and accurate MRI analysis. The images are pre-processed to increase the contrast and to remove the skull region. A novel algorithm is used to check whether the given image is normal or not. This algorithm reduces the computational complexity and increase the speed of proposed classification system by selecting abnormal images alone for further processing. Segmentation is performed on abnormal images to find the tumor region. Segmentation is based on a hybrid algorithm using K-means clustering and Texture Pattern Matrix. Texture Features and shape features are separately extracted from the segmented binary image using GrayLevel Co-occurrence Matrix (GLCM) and connected regions. The features thus obtained are used to train the neural network using Back Propagation Algorithm defined by Levenberg-Marquardt (LM) algorithm. Feed Forward Neural Network (FFNN) is used for the classification of MR images. While using the proposed method, accuracy is 98.06%, specificity is 97.77% and sensitivity is 98.34%. Speed, Robustness and computational complexity are the major advantages of the proposed system.
Totally 70 normal cases and 60 abnormal cases were analysed. It has been found from the Table 1- 4 that the correlation level is form 0.6 – 0.9, the entropy range is from 1.9 - 3.1 and the contrast value is around 1 which will enable the proposed classifier system to classify the nature of tissue with high accuracy. The accuracy of the classifier was evaluated based on the error rate. The error rate was calculated using the terms true positive (TP), true negative (TN), false positive (FP) and false negative (FN). Sensitivity and Specificity are the statistical measures used to analyse the performance of the classifier.
Abstract This paper shows the results obtained from images processing digitized, taken with a 'smartphone', of 56 samples of crushed olives, using the methodology of the gray-level co-occurrence matrix (GLCM). The values of the appropriate direction (θ) and distance (D) that two pixel with gray tone are neighbourhood, are defined to extract the information of the parameters: Contrast, Correlation, Energy and Homogeneity. The values of these parameters are correlated with several characteristic components of the olives mass: oil content (RGH) and water content (HUM), whose values are in the usual ranges during their processing to obtain virgin olive oil in mills and they contribute to generate different mechanical textures in the mass according to their relationship HUM / RGH. The results indicate the existence of significant correlations of the parameters Contrast, Energy and Homogeneity with the RGH and the HUM, which have allowed to obtain, by means of a multiple linear regression (MLR), mathematical equations that allow to predict both components with a high degree of correlation coefficient, r = 0.861 and r = 0.872 for RGH and HUM respectively. These results suggest the feasibility of textural analysis using GLCM to extract features of interest from digital images of the olives mass, quickly and non-destructively, as an aid in the decision making to optimize the production process of virgin olive oil.
Cholesterol is a waxy fat compound that is mostly produced by the liver and the other part is obtained from food. The ideal cholesterol level in the human body is <200. High cholesterol can increase the risk of getting serious diseases such as strokes and heart attacks. Checking cholesterol levels through checking blood sugar requires the patient to undergo fasting for 10-12 hours first and processing the results of the examination also requires not a short time. Because of the seriousness of the disease that can be caused, an early examination is needed and it is also practical to determine the level of excess cholesterol in the human body. Iris has specific advantages which can record all organ conditions, body construction and psychological conditions. Therefore, Iridology as a science based on the arrangement of the iris can be an alternative for medical analysis. In this study, the author designed a system in the matrix simulator which is expected to be able to detect excess cholesterol levels with input in the form of iris images and then through the pre-processing stage then extracted features with the GrayLevel Co-Occurrence Matrix method and classified using the Linear Regression method. The result from the modeling process can inform about cholesterol level. These processes make early detection of human body cholesterol level becomes easier. The cholesterol data level is classified into: normal cholesterol, at risk of cholesterol and high cholesterol. Each class was represented by 30 images, and each of it divided into two data types, 20 images used as training data and the remaining as testing data. The optimum result can be obtained on 45 degree angle, two pixels gap and correlation feture, which give 88.52% accuracy with 6.9595 standard deviation and 0.0365 seconds computation time for each image.
Normally brain tumor image contain large amount of information’s so manual segmentation is time consuming and complex process. In order to overcome these complication, automatic segmentations methods were introduced to detect the of brain tumor such as region growing , thresholding , artificial neural network  and clustering , etc. But still segmentation of brain tumor is still a challenging problem in image processing and analysis. The structure of the brain is complicated so it is difficult to determine the accurate segmentation of necrosis, edema and enhanced tumor. Several tissues present in the brain consist of three normal tissue  region, namely, gray matter (GM), white matter (WM), and Cerebrospinal Fluid (CSF), which is significant to analysis and treatment for diseases such as multiple sclerosis, Alzheimer's disease and epilepsy. These three regions are identified by the segmentation of brain image by utilizing the graylevel distribution of pixels. The main goal of brain tumor segmentation  is to identifies the extensive and location of the tumor region , such as edema, active tumorous tissue and necrotic tissue. Mean-shift algorithm  were used to detect the brain tumor in MRI image. The most widely used automatic segmentation technique in bioinformatics application  is clustering. Now a days, clustering based image segmentation on pixels are used in imaging technique, which organizing a given database into a group. Significant role of clustering in MRI image is generally used to detect the brain diseases and abnormalities, to monitor, diagnose and treat disease. Several clustering technique are used in the existing work to detect the abnormalities such as fuzzy k-mean clustering , , adaptive fuzzy k-mean clustering , modified k- mean clustering  and fuzzy C-mean  clustering. The goal of these clustering is to detect the abnormalities based algorithm to minimize the objective function based on certain criteria. In this paper, we introduce a Novel center ______________________________
In the field of computer vision, brain tumor classification has been widely practiced. Research on the classification of brain tumors can be divided into several phases, starting from the process of segmentation, extraction feature on the object area, until to build a model to recognize the type of brain tumor. Several approaches to classifying brain tumors have been performed , . In the case of image classification, the most commonly used method is Deep Learning which is considered to have a high degree of accuracy . One method of Deep Learning is the Deep Neural Network, two of which are the Convolutional Neural Network (CNN) and Multilayer Perceptron (MLP). CNN is the development of MLP. CNN is rated better than MLP because CNN has a high network depth and is widely applied to image data, whereas MLP is not good because it does not store spatial information from image data and assumes each pixel is an independent feature that produces poor results .
This study was approved by the institutional review board of our hospital, and informed consent was obtained from all patients. In this study, we recruited 369 patients with 446 lesions including HCC (222/446) and HH (224/446) confirmed by pathological examina- tions after hepatectomy, puncture assessments, interven- tional tumour cell staining or typical imaging findings between January 1, 2011 and September 1, 2017. A train- ing set was constituted by randomly selecting 80% of the samples and the remaining samples were used to test. All patients had undergone MRI examinations before therapeutic procedures such as surgery, puncture, and adjuvant therapy. MRI was performed using the GE 3 .0T or 1 .5T MR scanner (Signa, HDxt, GE Healthcare, United States) with an eight-channel phased array body coil. All patients were asked to fast for about four to six hours before scanning, after which they underwent upper abdomen MRI examination in the supine position. The MR scan sequences were as follows: (1) fast spoiled gradient-recalled sequence axial fat suppression T1-weighted gradient echo in-phase and out-phase: repetition time (TR)/echo time (TE) = 400/8.0 ms, field of view (FOV) = 320 mm × 320 mm, matrix = 320 × 192, number of excitations (NEX) = 2.0, slice thickness = 5.0 mm. (2) T2WI: TR/TE = 4000/125 ms, FOV = 320 mm × 320 mm, matrix = 320 × 192, NEX = 4.0, slice thickness = 5 .0mm. (3) DWI with a spin-echo planar imaging se- quence: TR/TE, 4000/70 ms, b = 600 s/mm , slice thickness = 5.0 mm, FOV = 320 mm × 320 mm, matrix = 128 × 128, NEX = 6.0. The patients ’ status of hepatic cir- rhosis were recorded according to MRI features of cir- rhosis: a nodular liver margin, lobar atrophy / hypertrophy, parenchymal heterogeneity. Finally, 59 HCC lesions and 19 HH lesions were accompanied by the occurance of cirrhosis.
failing in low contrast images. Based on statistical measures like mean, variance, standard deviation Otsu derived threshold. Shannon introduced information theory based on the concept of entropy . Pun used this entropy concept to derive threshold . Kapur et.al. improved the work of Pun . This is extended to Renyi’s entropy [8-9] and Tsalli’s entropy [10-11]. Yang Xiao et.al. Improved this work by constructing 3D GrayLevelSpatial Correlation (GLSC) histogram  by considering the local properties of image at constant similarity measure 4 which overcomes the time complexity of 2D histogram approaches . The usage of 3D histogram instead of 2D will result better threshold value, GrayLevelSpatial Correlation (GLSC) histogram along with entropic techniques is the recent advancement in this context. In this paper we propose an image segmentation technique based on GLSC histogram with dynamic similarity discrimination factor ( ) by considering the local and global characteristics, to improve the method proposed by Yang Xiao. The parent version algorithm using a constant 4 as the similarity measure to construct a 3D histogram on a 3×3 window image, does not suits for all types of images. Using Fuzzy technique to extract Fuzzyfied region  in image and calculating threshold using Shannon’s entropy  in this region itself makes the proposed image segmentation technique very time efficient . The redistribution of missing probability amount in floating precisions is made to improve the performance of the proposed method. The parameter efficiency based on misclassification error between segmented image and ground truth image is more than the existing methods.
Color is one of the most reliable visual features that are also easier to implement in image retrieval systems. Color is independent of image size and orientation, because, it is robust to background complication. Color histogram is the most common technique for extracting the color features of colored images [2-6]. Color histogram tells the global distribution of colors in the images. It involves low computation cost and it is insensitive to small variations in the image structure. However, color histogram hold two major shortcomings. They are unable to fully accommodate the spatial information, and they are not unique and robust. Two dissimilar images with similar color distribution produce very similar histograms. Moreover, similar images of same point of view carrying different lighting conditions create dissimilar histograms.
Abstrak. Analisa tekstur adalah satu sifat penting untuk mengenal pasti permukaan dan objek daripada imej perubatan dan pelbagai imej lain. Penyelidikan ini telah membangunkan sebuah algoritma untuk menganalisa tekstur dengan menggunakan imej perubatan dari echocardiography untuk mengenal pasti jantung yang disyaki mengalami myocardial infarction. Di sini penggabungan daripada teknik wavelet extension transform dan teknik graylevel co-occurrence matrix adalah dicadangkan. Di dalam penyelidikan ini wavelet extension transform digunakan untuk menghasilkan sebuah imej hampiran yang mempunyai resolusi yang lebih besar. Graylevel co-occurrence matrix yang dihitung untuk setiap sub-band digunakan untuk mencirikan empat sifat vektor: entropy, contrast, energy (angular second moment) dan homogeneity (invers difference moment). Pengklasifikasian yang digunakan di dalam penyelidikan ini adalah pengklasifikasian Mahalanobis distance. Kaedah yang telah dicadangkan diuji dengan data klinikal dari imej echocardiography untuk 17 orang pesakit. Untuk setiap pesakit, contoh tisu diambil daripada kawasan yang disyaki infarcted dan kawasan non-infarcted (normal). Untuk setiap pesakit, 8 bingkai imej yang dipisahkan oleh sela waktu tertentu di mana 5 kawasan normal dan 5 kawasan disyaki myocardial infarction berukuran 16 × 16 piksel akan dianalisa. Hasil pengklasifikasian telah dicapai dengan ketepatan 91.32%.
Feature extraction is important in the process of getting the meaningful characteristic or information used in classification [19,20,21]. Gray-level co-occurrence matrix (GLCM) is the technique to evaluate textures by considering the spatial relationship of the pixels. This method calculates the occurrence of pairs of pixels with specific values and in a specified spatial relationship in an image. The spatial relationship is means the pixel of interest and the pixel to its immediate right (horizontally adjacent). After that, it will produce the statistical method from the calculation matrix. The GLCM used in this experiment calculate the occurrence of gray-level value i in specific spatial relationship to a pixel of j and then sum the number of i appears in the specific spatial relationship to pixel with value of j in the image .
Since the efficiency of treatment of thyroid disorder depends on the risk of malignancy, indeter- minate follicular neoplasm (FN) images should be classified. The diagnosis process has been done by visual interpretation of experienced pathologists. However, it is difficult to separate the favor benign from borderline types. Thus, this paper presents a classification approach based on 3D nuclei model to classify favor benign and borderline types of follicular thyroid adenoma (FTA) in cytological specimens. The proposed method utilized 3D graylevel co-occurrence matrix (GLCM) and random forest classifier. It was applied to 22 data sets of FN images. Furthermore, the use of 3D GLCM was compared with 2D GLCM to evaluate the classification results. From experimental results, the proposed system achieved 95.45% of the classification. The use of 3D GLCM was better than 2D GLCM according to the accuracy of classification. Consequently, the proposed method probably helps a pathologist as a prescreening tool.