Search

The matching area selection algorithm is one of the key technologies for underwater gravity‐aided inertial navigation system, which directly affects the positioning accuracy and matching rate of underwater navigation. The traditional matching area selection algorithms usually use the statistical characteristic parameters of gravity field. However, the traditional algorithms are difficult to reflect the spatial relation characteristic of gravity field, which always miss some latent matching areas with obvious change of gravity field. In order to solve this problem, the matching area selection algorithm based on co‐occurrence matrix is proposed. The proposed algorithm establishes gravity anomaly co‐occurrence matrix and extracts spatial relation characteristic parameters to reflect the gravity field. The comprehensive spatial characteristic parameter is built by entropy and is used to select the matching area by maximization of inter‐class variance. The experimental results show that the proposed algorithm can select more effective matching areas than the traditional algorithms.

Nowadays, due to the increasing crime and theft around the world, surveillance security systems play an important role. On the other hand, the availability of video editing tools has made authenticity of video contents significant and urgent mission to use as strong evidence in the courts. Frame duplication with/without shuffling is a common form of video forgery to repeat or cover-up an event in a video’s scene. In this paper, we propose a robust method to detect inter-frame duplication forgery using a temporal average of each shot and statistical textural features. Duplicated shots containing frames that are reordered during the forgery process (frame shuffling), cannot be classified as tampered shots by the existing methods leading to an increase in false positives. To address this issue, we use a temporal average of each shot which found to be invariant with different orders. Our method is capable of detecting duplicate shots that do not have any tracing points (discontinuity points). Experimental results show that our method has achieved improved accuracy on frame duplication detection with lower computational time. Furthermore, it has successfully detected frame shuffling with high accuracy rates, even when the forged video has undergone post-processing operations such as Gaussian blurring, noise addition, brightness modification, and compression.

Surface Roughness of a machined component is crucial in identifying its functional capability when the manufactured specimen has metal to metal contact during operating condition since most wear and tear of the parts occurs due to friction between the surfaces of the moving parts. It is quite difficult to manually check the surface roughness of each component being manufactured on a manufacturing line. This paper aims to present a methodology to predict surface roughness using Image Processing, Computer Vision, and Machine Learning. Two machine learning algorithms Bagging Tree and Stochastic Gradient Boosting are compared and evaluated based on statistical parameters .It is observed that Stochastic Gradient Boosting predicts surface roughness in an efficient way both for training and Ten-fold cross-validation. The methodology used can be employed for online inspection and qualitative assessment of machined components.

Rapid boost in the density of the pedestrians and vehicles on the roads have made the life of visually impaired people very difficult. In this direction, we present the design of a smart phone based cost-effective system to guide visually impaired people to walk safely on the roads by detecting obstacles in real-time scenarios. Monocular vision based method is used to capture the video and then frames are extracted out of it after removing the blurriness caused by the motion of camera. For each frame, a computationally simple approach based on the ground plane is proposed for detecting and removing the ground plane. After removing ground plane, features like Speeded-Up Robust Features (SURF) of the non-ground area are computed and compared with features of obstacles. An active contour model is used to segment the area of non-ground image whose SURF features are matched with obstacle features. This area is referred as Region of Interest (ROI). To check whether ROI belongs to an obstacle or not, Gray Level Co-occurrence matrix (GLCM) features are calculated and passed onto a classification model. Classification results show that this system is efficiently able to detect the obstacles that are known to the system in near real-time.

Tracking objects in infrared video sequences became a very important challenge for many current tracking algorithms due to several complex situations such as illumination variation, night vision, and occlusion. This study proposes a new tracker that uses a set of invariant parameters calculated via the co-occurrence moments to better describe the target object. The usage of the co-occurrence moments gives the ability to exploit the information about the texture of the target to enhance the robustness of the tracking task. This latter is performed without any learning or clustering phase. The qualitative and quantitative studies on challenging sequences demonstrate that the results obtained by the proposed algorithm are very competitive in comparison to several state-of-the-art methods.

To improve the efficiency of blur kernel estimation based on prior knowledge, a method of deblurring an image based on rich edge region extraction using a gray-level co-occurrence matrix is proposed in this paper. First, the relationship between the image edge information and the related coefficients of a gray-level co-occurrence matrix is analyzed, based on which an index representing the amount of image edge information is proposed. Next, high-frequency layer information is extracted from the blurred image to be processed with a bilinear interpolation method in the luminance channel. Subsequently, the high-frequency layer image is divided into nine regions, based on a sliding window, and the rich edge region index of each region is calculated; then, the region with the richest edge information is extracted. Finally, the extracted rich edge region, instead of the entire motion blurred image, is used to estimate the blur kernel with L0-regularized intensity and gradient prior, and the blurred image is blindly restored. An image quality evaluation function and the operation time are used to evaluate the performance of the proposed method. Experimental results show that the proposed method can improve the recovery efficiency while ensuring the recovery quality as well.

Various fields in medicine require scientific research and computer application. This results in computation time optimization becoming a task that is of increasing importance due to its highly parallel architecture. As is well-known, the graphics processing unit (GPU) is regarded as a powerful engine for application programs that demand fairly high computation capabilities. Our study is based on the deep analysis of the parallelism pertaining to the calculation of the gray level co-occurrence matrix, whereby an algorithm was introduced to optimize the method used to compute the gray-level co-occurrence matrix (GLCM) of an image. Furthermore, strategies (e.g., copying, image partitioning, and so on) were proposed to optimize the parallel algorithm. Our experiments indicate that without losing the computational accuracy, the speed-up ratio of the GLCM computation of images with different resolutions by GPU utilizing compute unified device architecture was at least 50 times faster than that of the GLCM computation by the central processing unit. This manifestation of a significantly improved performance can lead to the development of a very useful computational tool in medical computer vision.

Texture analysis is devised to address the weakness of color-based image segmentation models by considering the statistical and spatial relations among the group of neighbor pixels in the image instead of relying on color information of individual pixels solely. Due to decent performance of the gray-level co-occurrence matrix (GLCM) in texture analysis of natural objects, this study employs this technique to analyze the human skin texture characteristics. The main goal of this study is to investigate the impact of major GLCM parameters including quantization level, displacement magnitudes, displacement direction and GLCM features on skin segmentation and classification performance. Each of these parameters has been assessed and optimized using an exhaustive supervised search from a fairly large initial feature space. Three supervised classifiers including Random Forest, Support Vector Machine and Multilayer Perceptron have been employed to evaluate the performance of the feature space subsets. Evaluation results using Edith Cowan University (ECU) dataset showed that the proposed texture-assisted skin detection model outperformed pixelwise skin detection by significant margin. The proposed method generates an F-score of 91.98, which is satisfactory, considering the challenging scenario in ECU dataset. Comparison of the proposed texture-assisted skin detection model with some state-of-the-art skin detection models indicates high accuracy and F-score of the proposed model. The findings of this study can be used in various disciplines, such as face recognition, skin disorder and lesion recognition, and nudity detection.

As the interest in one’s appearance has recently increased, the demand for diagnosing skin conditions has also increased. However, conventional specialized skin diagnostic devices are generally expensive, and people have to visit a skin-care shop to diagnose their skin condition. This is time consuming and troublesome. In this paper, we propose a skin-roughness estimation method that uses a mobile-phone camera in daily environments. In order to achieve accurate evaluation, the illumination variation is alleviated using texture components of the facial skin image. We also propose a new feature-extraction method based on the gray-level co-occurrence matrix, which effectively measures the skin roughness from the texture components. The performance of the proposed method is compared with the conventional commonly used features, and we verify the superiority of the proposed method.

Among the various brain diseases, stroke is the major cause of death worldwide, next to heart attack. This paper proposes an algorithm in predicting the ischaemic stroke lesion using midline sketching and histogram bin-based technique. The visible ischaemic stroke lesion region and the normal region of the same computed tomography image are segmented with the help of histogram bins and the features are extracted. The first- and second-order statistical features for both regions are analysed. The differences in the features are utilised to categorise the lesion and non-lesion region. The statistical t-test analysis-based observations with a confidence interval of 95% for each feature are tabulated. These observations indicate that among the nine features, as per the statistical analysis, six features provide the clear differentiation between normal and abnormal regions.

Crop classification in early phenological stages has been a difficult task due to spectrum similarity of different crops. For this purpose, low altitude platforms such as drones have great potential to provide high resolution optical imagery where Machine Learning (ML) applied to classify different types of crops. In this research work, crop classification is performed at different phenological stages using optical images which are obtained from drone. For this purpose, gray level co-occurrence matrix (GLCM) based features are extracted from underlying gray scale images collected by the drone. To classify the different types of crops, different ML algorithms including Random Forest (RF), Naive Bayes (NB), Neural Network (NN) and Support Vector Machine (SVM) are applied. The results showed that the ML algorithms performed much better on GLCM features as compared to gray scale images with a margin of 13.65% in overall accuracy.

Color texture classification as a part of fabric analysis is significant for textile manufacturing. In this research, a new artificial intelligence method based on a dual-side co-occurrence matrix and a back propagation neural network has been proposed for color texture classification, which could achieve relatively accurate classification results for yarn-dyed woven fabric compared with the traditional co-occurrence matrix for a single-side image. Firstly, a laboratory dual-side imaging system has been established to digitize the upper-side and lower-side images sequentially. Secondly, the dual-side co-occurrence matrix could be generated based on these dual images; four texture features could be extracted for the evaluation of the fabric texture characteristics. Thirdly, a well-trained back propagation neural network was established with the four defined features as the input vectors and the color texture type of yarn-dyed woven fabric as the output vector. The efficiency of two different classification systems based on a dual-side co-occurrence matrix and a single-side co-occurrence matrix has been compared systematically. Our experimental results show that the artificial intelligence system based on a dual-side co-occurrence matrix and back propagation neural network model could achieve a relatively better classification effect, with the high coefficient ratio ( R = 0.9726) when d = 0.

Cutting and extruding processing technology for ceramics based on the edge-chipping effect is a non-traditional rough machining method for engineering ceramics. A set of new methods for evaluating unconventional rough surfaces of such ceramics was developed by using grey-level co-occurrence matrix (GLCM) and a neural network (NN). The influences of three parameters including step size, greyscale quantisation and direction on the GLCM were investigated to measure the morphology of the machined surface of Si3N4 ceramic by using a GLCM with suitable such parameters. Based on a generalised regression network, a prediction model for the textural features of sintered Si3N4 ceramic surfaces was established with multiple processing parameters. Moreover, a competitive layer network was used to sort the roughness grades of the machined surface. The division and cooperation of the generalised regression network and competitive network are able to preferably identify and predict the roughness of the machined surface without contact.

The vehicle with harmful black smoke pollutant emitted from vehicle exhaust pipe is usually called smoky vehicle. Existing smoky vehicle detection methods mainly lie on traditional manual monitoring. In this paper, we propose an intelligent smoky vehicle detection method based on Gray Level Co-occurrence Matrix (GLCM). This method can automatically detect smoky vehicles through analyzing the road surveillance videos. More specifically, we adopt Vibe background subtraction algorithm to detect vehicle objects. The gray-level integral projection technology and image local range technology are combined to detect the vehicle rear. We extract GLCM from the region at the back of the vehicle, and five different GLCM-based features, namely, angular second moment (ASM), entropy (ENT), contrast (CON), correlation (COR), and inverse difference moment (IDM), are selected to distinguish smoky images and nonsmoke images. The back propagation (BP) neural network is adopted to train the classifier and classify new samples. The experimental results show that the proposed method has a good performance.

The change of crowd energy is a fundamental measurement for describing a crowd behavior. In this paper, we present a crowd abnormal detection method based on the change of energy-level distribution. The method can not only reduce the camera perspective effect, but also detect crowd abnormal behavior in time. Pixels in the image are treated as particles, and the optical flow method is adopted to extract the velocities of particles. The qualities of different particles are distributed as a different value according to the distance between the particle and the camera to reduce the camera perspective effect. Then a crowd motion segmentation method based on flow field texture representation is utilized to extract the motion foreground, and a linear interpolation calculation is applied to pedestrian’s foreground area to determine their distance to the camera. This contributes to the calculation of the particle qualities in different locations. Finally, the crowd behavior is analyzed according to the change of the consistency, entropy and contrast of the three descriptors for co-occurrence matrix. By calculating a threshold, the timestamp when the crowd abnormal happens is determined. In this paper, multiple sets of videos from three different scenes in UMN dataset are employed in the experiment. The results show that the proposed method is effective in characterizing anomalies in videos.

Fingerprint-based recognition systems have been widely deployed in numerous civilian and government applications. However, the fingerprint recognition systems can be deceived by commonly used sensors with the artificially fake fingerprint made using materials like gelatin or silicon. In this paper, spoof fingerprint detection is considered as a two-class classification problem and co-occurrence matrix is constructed from image gradients to extract features. In feature extraction process, the quantization operation is firstly applied with the fingerprint images. Secondly, the horizontal and vertical differences at each pixel are calculated. Thirdly, the differences of large absolute values are truncated into a reduced range. Finally, the co-occurrence matrix is constructed from the truncated differences, and the elements of the co-occurrence matrix are directly used as features. The features are separately utilized to train support vector machine classifiers on two databases. The experimental results have demonstrated that the proposed method outperform the state-of-the-arts.

This research investigates detection and classification of two types of the surface defects in extruded aluminium profiles; blisters and scratches. An experimental system is used to capture images and appropriate statistical features from a novel technique based on gradient-only co-occurrence matrices (GOCM) are proposed to detect and classify three distinct classes; non-defective, blisters and scratches. The developed methodology makes use of the Sobel edge detector to obtain the gradient magnitude of the image (GOCM). A comparison is made between the statistical features extracted from the original image (GLCM) and those extracted from the gradient magnitude (GOCM). This paper describes in detail every step of the image processing with example pictures illustrating the methodology. The features extracted from the image processing are classified by a two-layer feed-forward artificial neural network. The artificial neural network training is tested using different combinations of statistical features with different topologies. Features are compared individually and grouped. Results are discussed, achieving up to 98.6 % total testing accuracy.

Texture attributes describe the spatial arrangement of neighboring amplitudes values within a given analysis window. We chose a statistical texture classification method, the gray-level co-occurrence matrix (GLCM), and its derived attributes, to produce a semiautomated description of the spatial arrangement of seismic facies. The GLCM is a measure of how often different combinations of neighboring pixel values occur. We tested the application of directional GLCM-based attributes for the detection of seismic variability within paleoriver features. Calculation of 3D GLCM-based attributes can be done in 13 space directions. The results of GLCM-based attribute calculation differed depending on the chosen GLCM parameters (number of gray levels, analysis window, and direction of calculation). We specifically focused on how the direction of calculation influenced the computation of attributes, while keeping other parameters constant. We first tested the workflow on a 2D training image and later ran on a real seismic amplitude volume from the Vienna Basin. Based on the GLCM-based attributes, we could map the channel features and extract them as geobodies. Additionally, we generated a new set of directional GLCM-based attributes to detect spatial changes in the seismic facies. By comparing these directional attributes, we could determine areas within the channel features having higher directional variability. Areas with higher tendency to directional variations might be associated with changes in lithology, seismic facies, or with seismic anisotropy.

The pattern identification problems such as stone, rock categorization and wood recognition are used texture classification technique due to its valuable usage in it. Generally, texture analysis can be done one of the two ways i.e. statistical and structural approaches. More problems are occurred when working with statistical approaches in texture analysis for texture categorization. One of the most popular statistical approaches is Gray Level Co-occurrence Matrices (GLCM) approach. This approach is used to discriminating different textures in images. This approach gives better accuracy results but this takes high computational cost. Usually, texture analysis method depends upon how the texture features are extracted from the image to characterize image. Whenever a new texture feature is derived it is tested whether it is precisely classifies the textures or not. Texture features are most important for precise and accurate texture classification and also important that the way in which they are extracted and applied. The present paper derived a new co-occurrence matrix based on overlapped textons patterns. The present paper generates overlapped texton patterns and generates co-occurrence matrices derived a new matrix called Overlapped Texton Co-occurrence Matrices (OTCoM) for stone texture classification. The present paper integrates the advantages of co-occurrence matrix and texton image by representing the attribute of co-occurrence. The co-occurrence features extracted from the OTCoM provides complete texture information about a texture image. The proposed method is experimented on Vistex, Brodatz textures, CUReT, Mayang, Paul Brooke, and Google color texture images. The experimental results indicate the proposed method classification performance is superior to that of many existing methods.

The characteristic value of gray level co-occurrence matrix to extract can well express the information of texture. Co-occurrence matrix provides the information of image grayscale, interval and change. According to the co-occurrence matrix, it can calculate the corresponding characteristic values of eigenvalue, which can express the texture information of the image. This is thesis designed extraction software a for textile fabric texture feature, and the internal principle is the using of gray level co-occurrence matrix and Matlab programming.

The human eye can easily identify the type of textures in flooring of the houses and in the digital images visually. In this work, the stone textures are grouped into four categories. They are bricks, marble, granite and mosaic. A novel approach is developed for decreasing the dimension of stone image and for reducing the gray level range of the image without any loss of significant feature information. This model is named as “Decreased Dimension and Reduced Gray level Range Matrix (DDRGRM)” model. The DDRGRM model consists of 3 stages. In stage 1, each 5×5 sub dimension of the stone image is reduced into 2×2 sub dimension without losing any important qualities, primitives, and any other local stuff. In stage 2, the gray level of the image is reduced from 0-255 to 0-4 by using fuzzy concepts. In stage 3, Co-occurrence Matrix (CM) features are derived from the DDRGRM model of the stone image for stone texture classification. Based on the feature set values, a user defined algorithm is developed to classify the stone texture image into one of the 4 categories i.e. Marble, Brick, Granite and Mosaic. The proposed method is tested by using the K-Nearest Neighbor Classification algorithm with the derived texture features. To prove the efficiency of the proposed method, it is tested on different stone texture image databases. The proposed method resulted in high classification rate when compared with the other existing methods.

Retrieving images depend of specific features In Content Based Image Retrieval (CBIR). A common approach is to divide retrieval process into two stages; the first one is based on high-level features followed by the second that is based on low-level features. We focus primarily on medical images, and follow the above approach but make the following two basic contributions: a) introduce the gray cluster co-occurrence matrix as texture feature extraction and use it as high-level features, and b) introduce edge strength levels as shape feature extraction and use it as low-level features. Our proposed system suggests the precision rate was 94.90% and recall rate was 89.72%. The distance variance achieved lowest rate (0.0022) in images retrieval compared to each of partial systems individually and related works. Our method has better performance in retrieving the results than other related works and each of partial system individually.

The rapid growth of different types of images has posed a great challenge to the scientific fraternity. As the images are increasing everyday, it is becoming a challenging task to organize the images for efficient and easy access. The field of image retrieval attempts to solve this problem through various techniques. This paper proposes a novel technique of image retrieval by combining Scale Invariant Feature Transform (SIFT) and Co-occurrence matrix. For construction of feature vector, SIFT descriptors of gray scale images are computed and normalized using z-score normalization followed by construction of Gray-Level Co-occurrence Matrix (GLCM) of normalized SIFT keypoints. The constructed feature vector is matched with those of images in database to retrieve visually similar images. The proposed method is tested on Corel-1K dataset and the performance is measured in terms of precision and recall. The experimental results demonstrate that the proposed method outperforms some of the other state-of-the-art methods.

The development of moving object detection systems has become of great interest in the video surveillance field. Although many foreground detection methods have been proposed, the problems of incomplete object shapes and misclassified shadow regions remain. We propose a new approach by combining bit-planes representation with gray-level co-occurrence matrix (GLCM). This allows the system to exploit the motion history and remove the shadow, respectively. As shown by experiments in Section 3, our approach is efficient and robust for detecting moving objects in real-time.

This paper presents a comparative study to the most common texture features analysis methods. In fact, there are two kinds of approaches have been proposed to extract the texture features for the purpose of object categorization, the former deals with the intensity pixels which derived the intensities texture and the second method dealing with the edge pixels which obtained the edge texture. However, to extract and make a comparative analysis to the texture maps there are several approaches have been presented in this research, i.e., Gradient, Co-occurrence matrix, Contrast and Edge Histogram Descriptor. A real world images dataset denoted by Caltech 101 dataset has been adopted to evaluate the proposed texture analysis methods. Mostly, the first 20 classes with 40 images per-class have been chosen to demonstrate the methods performance. Fundamentally, the objects of these images have almost isolated for the purpose of categorization. The experiment results show that the edge histogram descriptor outperformed the other proposed texture analysis methods with average accuracy 71.175±1.355775 because the edge histogram descriptor is less sensitive to the noise and variation of pixels intensities which most of the objects of Caltech 101 dataset profoundly affected.

Focal nodular hyperplasia (FNH), hepatocellular carcinoma (HCC) and cavernous hemangioma (HEM) are three types of solid focal liver lesions. Using CT images to identify these three types of lesions is the most commonly method. However, this method usually mainly depends on experiences. Whereas, more objective and quantitative image information could be explored with texture analysis method. This research aims to discuss the appreciations of texture analysis based on CT images for identifying FNH, HCC and HEM. This paper retrospectively analyzed 81 clinically or pathologically diagnosed cases, each of which contains contrast non-enhanced and contrast two-phasic enhanced CT images. The texture analysis was based on gray level co-occurrence matrix (GLCM) and wavelet transform. The results shows that the misclassification rates of texture classification were low to 2.73% between FNH and HEM (between benign lesions), 3.19% between FNH and HCC (between benign lesions and malignant lesions), and 1.67% between HEM and HCC (between benign lesions and malignant lesions) respectively. The effect of texture classification based on contrast two-phasic enhanced CT images was better than contrast non-enhanced CT images.

Gray Level Co-occurrence Matrix (GLCM) is one of the main techniques for texture analysis that has been widely used in many applications. Conventional GLCMs usually focus on two-dimensional (2D) image texture analysis only. However, a three-dimensional (3D) image volume requires specific texture analysis computation. In this paper, an extended 2D to 3D GLCM approach based on the concept of multiple 2D plane positions and pixel orientation directions in the 3D environment is proposed. The algorithm was implemented by breaking down the 3D image volume into 2D slices based on five different plane positions (coordinate axes and oblique axes) resulting in 13 independent directions, then calculating the GLCMs. The resulted GLCMs were averaged to obtain normalized values, then the 3D texture features were calculated. A preliminary examination was performed on a 3D image volume (64 x 64 x 64 voxels). Our analysis confirmed that the proposed technique is capable of extracting the 3D texture features from the extended GLCMs approach. It is a simple and comprehensive technique that can contribute to the 3D image analysis.

Batik is one of Indonesian’s traditional cloth. Motif or pattern drawn on a piece of batik fabric has a specific name and philosopy. Although batik cloths are widely used in everyday life, but only few people understand its motif and philosophy. This research is intended to develop a batik motif recognition system which can be used to identify motif of Batik image automatically. First, a batik image is decomposed into sub-images using wavelet transform. Six texture descriptors, i.e. max probability, correlation, contrast, uniformity, homogenity and entropy, are extracted from gray-level co-occurrence matrix of each sub-image. The texture features are then matched to the template features using canberra distance. The experiment is performed on Batik Dataset consisting of 1088 batik images grouped into seven motifs. The best recognition rate, that is 92,1%, is achieved using feature extraction process with 5 level wavelet decomposition and 4 directional gray-level co-occurrence matrix.

In this paper, we present a method for the detection of fruits from images. Several fruits detection techniques have been developed based on colour and shape values. However, different object images may have similar or identical colour and shape. Therefore, using only colour and shape features are not efficient enough for fruits detection. Considering this fact, in this paper, we propose a method using colour, shape and texture for detecting objects from images. We performed several experiments using fruits images and found that our system can almost accurately detect fruits types from images.

Gray-level co-occurrence matrix (GLCM) is a statistical method widely used to characterize images and specifically, for Second Harmonic Generation (SHG) collagen images characterization. This method takes into account the spatial relationship between the image pixels, at specific angle. It is usually calculated for four orientations, at specific distances. Over these matrix, a textural feature function is calculated. Often, results of different orientations are compared or averaged to get a unique statistic parameter. In the present report, we will demonstrate the error that bring with this methodology, and following, we offer the correction formula. Preferred orientation of SHG images is proposed as structural property to characterize biological samples. For example, for determining the parallelism grade of collagen fibers regarding the ovarian epithelium. Here, we present a robust method to calculate this parameter, based on the two-dimensional Fourier transform. Finally, we show how these two elements help improve the discrimination between normal and pathological ovarian tissues.

Indonesian Batik has been endorsed as world cultural heritages by UNESCO. The batik consists of various motifs each of whom represents characteristics of each Indonesian province. One of the motifs is called as Batik Kain Besurek or shortly Batik Besurek, originally from Bengkulu Province. This motif constitutes a motif family consisting of five main motifs: Kaligrafi, Rafflesia, Relung Paku, Rembulan, and Burung Kuau. Currently most Batik Besureks reflect a creation developed from combination of main motifs so that it is not easy to identify its main motif. This research aims to classify Indonesian batik according to its image into either batik besurek or not batik besurek as well as reidentifying its more detailed motif for the identified batik besurek. The classification is approached through six classes: five classes in accordance with classification of Batik Besurek and a class of not Batik Besurek. The preprocessing system converts images to grayscale and followed by resizing. The feature extraction uses GLCM method yielding six features and SURF method yielding 64 descriptors. The extraction results are combined by assigning weight on both methods in which the weighting scheme is tested. Moreover, the image classification uses a method of k-Nearest Neighbor. The system is tested through some scenarios for the feature extraction and some values k in k-NN to classify the main motif of Batik Besurek. So far the result can improve system performance with an accuracy of 95.47% according to weighting 0.1 and 0.9 for GLCM and SURF respectively, and k = 3.

This paper presents the integration of color histogram and DBC co-occurrence matrix for content based image retrieval. The exit DBC collect the directional edges which are calculated by applying the first-order derivatives in 0o , 45o , 90o and 135o directions. The feature vector length of DBC for a particular direction is 512 which are more for image retrieval. To avoid this problem, we collect the directional edges by excluding the center pixel and further applied the rotation invariant property. Further, we calculated the co- occurrence matrix to form the feature vector. Finally, the HSV color histogram and the DBC co-occurrence matrix are integrated to form the feature database. The retrieval results of the proposed method have been tested by conducting three experiments on Brodatz, MIT VisTex texture databases and Corel-1000 natural database. The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP, DBC and other transform domain features.

This paper firstly studies the texture features. We construct a gray-difference primitive co-occurrence matrix to extract texture features by combining statistical methods with structural ones. The experiment results show that the features of the gray-difference primitive co-occurrence matrix are more delicate than the traditional gray co-occurrence matrix.

Concrete is an important material in most of civil constructions. Many properties of concrete can be determined through analysis of concrete images. Image segmentation is the first step for the most of these analyses. An automated system for segmentation of concrete images into microstructures using texture analysis is proposed. The performance of five different classifiers has been evaluated and the results show that using an Artificial Neural Network classifier is the best choice for an automatic image segmentation of concrete.

This study proposes a novel method for multichannel image gray level co-occurrence matrix (GLCM) texture representation. It is well known that the standard procedure for the automatic extraction of GLCM textures is based on a mono-spectral image. In real applications, however, the GLCM texture feature extraction always refers to multi/hyperspectral images. The widely used strategy to deal with this issue is to calculate the GLCM from the first principal component or the panchromatic band, which do not include all the useful information. Accordingly, in this study, we propose to represent the multichannel textures for multi/hyperspectral imagery by the use of: (1) clustering algorithms; and (2) sparse representation, respectively. In this way, the multi/hyperspectral images can be described using a series of quantized codes or dictionaries, which are more suitable for multichannel texture representation than the traditional methods. Specifically, K-means and fuzzy c-means methods are adopted to generate the codes of an image from the clustering point of view, while a sparse dictionary learning method based on two coding rules is proposed to produce the texture primitives. The proposed multichannel GLCM textural extraction methods were evaluated with four multi/hyperspectral datasets: GeoEye-1 and QuickBird multispectral images of the city of Wuhan, the well-known AVIRIS hyperspectral dataset from the Indian Pines test site, and the HYDICE airborne hyperspectral dataset from the Washington DC Mall. The results show that both the clustering-based and sparsity-based GLCM textures outperform the traditional method (extraction based on the first principal component) in terms of classification accuracies in all the experiments.

SUMMARY The necessity of distinguishing the road surface condition at night time is increasing because most of the previously proposed methods correspond only to daytime. In this paper, we propose a method of distinguishing road surface condition using only video information acquired from a visible surveillance video camera. The feature of this method is that it simply uses automobile headlights as the light source. Thus it becomes possible to distinguish between road surface conditions, such as dry and wet, with high accuracy using this method.

Vibrations are one of the obstacles to productivity of machining process since their presence reduces surface quality, dimensional accuracy and tool life. This article proposes a vision-based approach for determining vibration level in metal cutting. Vibration level of cutting tool is controlled by changing the tool overhang, and the resulting irregularity of surface texture is used as a criterion for determining the cutting tool vibration. Undecimated wavelet transform is used to decompose the surface image of the workpiece into sub-images in which the cutting tool vibration can be indicated. The texture of the preferred sub-image is analyzed using gray-level co-occurrence matrix texture features. In order to validate the proposed vision-based method, an accelerometer was attached to the shank of the cutting tool to measure vibrations in tangential direction. The experimental results showed that the combination of undecimated wavelet decomposition and gray-level co-occurrence matrix texture features can be used as a robust method for determining vibration level in the turning process.