Search

We present a new self-supervised monocular depth estimation method with multi-scale texture detail enhancement. Based on the observation that the image texture detail and the semantic information have essential significance on the depth estimation, we propose to provide them to the network to learn more sharpness and structural integrity of depth. Firstly, we generate the filtered images and detail images by multi-scale decomposition and use a deep neural network to automatically learn their weights to construct the texture detail enhanced image. Then, we consider the semantic features by putting deep features from the VGG-19 network into a self-attention network, guide the depth decoder network to focus on the integrity of objects in the scene. Finally, we propose a scale-invariant smooth loss to improve the structural integrity of the predicted depth. We evaluate our method on the KITTI 2015 and Make3D datasets and apply the predicted depth to novel view synthesis. The experimental results show that it has achieved satisfactory results compared with the existing methods.

To inspect the quality of concrete structures, surface voids or bugholes existing on a concrete surface after the casting process needs to be detected. To improve the productivity of the inspection work, this study develops a hybrid intelligence approach that combines image texture analysis, machine learning, and metaheuristic optimization. Image texture computations employ the Gabor filter and gray-level run lengths to characterize the condition of a concrete surface. Based on features of image texture, Support Vector Machines (SVM) establish a decision boundary that separates collected image samples into two categories of no surface void (negative class) and surface void (positive class). Furthermore, to assist the SVM model training phase, the state-of-the-art history-based adaptive differential evolution with linear population size reduction (L-SHADE) is utilized. The hybrid intelligence approach, named as L-SHADE-SVM-SVD, has been developed and complied in Visual C#.NET framework. Experiments with 1000 image samples show that the L-SHADE-SVM-SVD can obtain a high prediction accuracy of roughly 93%. Therefore, the newly developed model can be a promising alternative for construction inspectors in concrete quality assessment.

Fractal image compression coding algorithm is a novel image compression technology; however, the long encoding time and unacceptable image reconstruction quality remain the primary obstacles in practical application. The purpose of this study is to improve the coding quality from the perspective of grey level transform and feature extraction. In this study, a novel orthogonal sparse fractal coding algorithm based on image texture feature is proposed. The authors define a normalised version as the new grey description feature of the image block so that two improved methods are scientifically combined in theory and algorithm. First, orthogonal sparse grey level transform based on sparse decomposition improves image reconstruction quality and decoding speed. Then, the similarity measure matrix, which stores the variance feature between range blocks and domain blocks, is used to reduce redundancies and encoding time. Simulation results show that the proposed algorithm in this study can obtain better image reconstruction quality and speed up encoding time significantly as compared to the conventional fractal coding schemes.

Automatic classification in medical image processing with new problems to computer-aided detection methods is attributed to the dynamics of the image acquisition conditions. Such problems request t...

Color constancy is the capability to observe the true color of a scene from its image regardless of the scene's illuminant. It is a significant part of the digital image processing pipeline and is utilized when the true color of an object is required. Most existing color constancy methods assume a uniform illuminant across the whole scene of the image, which is not always the case. Hence, their performances are influenced by the presence of multiple light sources. This paper presents a color constancy adjustment technique that uses the texture of the image pixels to select pixels with sufficient color variation to be used for image color correction. The proposed technique applies a histogram-based algorithm to determine the appropriate number of segments to efficiently split the image into its key color variation areas. The K-means++ algorithm is then used to divide the input image into the pre-determined number of segments. The proposed algorithm identifies pixels with sufficient color variation in each segment using the entropies of the pixels, which represent the segment's texture. Then, the algorithm calculates the initial color constancy adjustment factors for each segment by applying an existing statistics-based color constancy algorithm on the selected pixels. Finally, the proposed method computes color adjustment factors per pixel within the image by fusing the initial color adjustment factors of all segments, which are regulated by the Euclidian distances of each pixel from the centers of gravity of the segments. The experimental results on benchmark single- and multiple-illuminant image datasets show that the images that are obtained using the proposed algorithm have significantly higher subjective and very competitive objective qualities compared to those that are obtained with the state-of-the-art techniques.

The traditional image texture extraction algorithm, most of the lifting of the texture is a single scale and direction of the texture. When you change the scale and direction, the corresponding texture will change as well. This is the lack of the traditional texture extraction method. In this paper, aiming at the single direction of the traditional texture extraction, a novel image texture feature extraction algorithm based on Gabor filter and CS-LBP operator is proposed. The extracted frequency components are transformed to the time domain to achieve the texture extraction. Based on the principle of polarization, the curvature factor is introduced based on the traditional Gabor model, which solves the problem that the Gabor filter can’t extract the local image warp and distant scene blurred. The experimental results show that the proposed method uses the Gabor filter and the LBP operator to extract the polarized image texture effectively than the traditional image texture extraction algorithm.

International audience; The importance of image compression is now essential during transmission or storage processes in various data applications, especially in medical and biometric systems. To perform the effectiveness of the compression process on images and evaluate degradation caused by this process, image quality assessment becomes an important tool in image services. We note that the objective criteria in image quality depend especially on the image type and image texture composition. The actual tendency is to find metrics making better qualification on errors in compressed images and correlate with the human visual system. This paper presents an investigation to examine and evaluate image compression degradation by the use of a new tendency concept of image quality assessment based on texture and edge analysis. To perform and practice this evaluation, we compress the medical and biometric images using second-generation wavelet compression algorithms and study the degradation of texture information in these images.

Digital watermarking is a key technology to solve copyright protection and content authentication. Most existing watermarking algorithms are based on global embedding, which cannot well balance the imperceptibility and robustness of watermarking. This paper proposes an adaptive robust watermarking algorithm based on image texture, which mainly includes: (1) color image is converted from RGB space to Lab space and the stable scale invariant feature transform(SIFT) points are extracted based on L component as the embedding position of watermark; (2)the structured forest edge is extracted using machine learning as the watermark image which is decomposed by using lifting wavelet transform (LWT) and then encrypted using logical chaos transform; (3)in consideration of human visual system, the strength factor is adaptively selected by using the brightness information and texture complexity of the L component. Experimental results show that the proposed algorithm in Lab space has the better visual invisibility and robustness to resist various attacks, especially for cropping, noise and JPEG compression attacks in comparison with other related algorithms.

Feature extraction is a key step in image processing for pattern recognition and machine learning processes. Its purpose lies in reducing the dimensionality of the input data through the computing of features which accurately describe the original information. In this article, a new feature extraction method based on Discrete Modal Decomposition (DMD) is introduced, to extend the group of space and frequency based features. These new features are called modal features. Initially aiming to decompose a signal into a modal basis built from a vibration mechanics problem, the DMD projection is applied to images in order to extract modal features with two approaches. The first one, called full scale DMD, consists in exploiting directly the decomposition resulting coordinates as features. The second one, called filtering DMD, consists in using the DMD modes as filters to obtain features through a local transformation process. Experiments are performed on image texture classification tasks including several widely used data bases, compared to several classic feature extraction methods. We show that the DMD approach achieves good classification performances, comparable to the state of the art techniques, with a lower extraction time., 12 pages, 5 figures. Accepted and to be published in Pattern Recognition Letters

In this paper, a scheme of facial texture synthesis and 3D reconstruction based on 2D face image and 3D model under the Unity platform is proposed. First of all, this scheme uses the ASM algorithm to extract the key points of the human face. Next, the facial features in the face image and the standard skin color image are texture blended to get the texture of fused facial image. Finally, the texture of fused facial image is mapped to the 3D face model by texture mapping, and a better personalized 3D face reconstruction effect is obtained by this method. Compared with FaceGen, a software for modeling facial photos, the experimental results show that the proposed algorithm is superior to other existing algorithms in efficiency and has certain application value.

Ice storms can cause major power disruptions and detection of ice formation on power cables can avoid them by taking preventive actions such as removing the ice before a major problem occurs. In this paper, a computer vision solution was developed to detect ice on difficult imaging scenarios where illumination and variety of locations can change the image of the cable and background considerably. The methodology starts with edge detection and support vector regression was applied to predict the upper threshold for the Canny edge detector. We found that this improves the later classification accuracy. A total of 44 image features based on the gray level co-occurrence matrix, statistical features and Hough transform were extracted and then processed by principal component analysis and sequential forward selection to reduce the feature dimension. Final detection was performed using four different classifiers. It was found that eight features achieved optimal performance using a support vector machine.

In the flotation process, the flotation froth texture is an indicator of the flotation state. To recognize the flotation state based on texture features accurately and to provide guidance for production operations, this paper proposes a method for flotation froth image texture extraction based on the deterministic tourist walks algorithm. First, a weighted graph model of a froth image is built using deterministic tourist walks. Next, the degree distribution and the unit intensity distribution of the weighted graph are extracted. The contrast of the node degree and the contrast of the node unit intensity are calculated as the texture feature indexes. The texture feature indexes are used for flotation production state classification and recognition. The experimental results demonstrate that the proposed method can extract froth image texture features accurately and provide effective guidance for flotation production.

Strawberry is an important fruit crop in Canada but powdery mildew (PM) results in about 30&ndash, 70% yield loss. Detection of PM through an image texture-based system is beneficial, as it identifies the symptoms at an earlier stage and reduces labour intensive manual monitoring of crop fields. This paper presents an image texture-based disease detection algorithm using supervised classifiers. Three sites were selected to collect the leaf image data in Great Village, Nova Scotia, Canada. Images were taken under an artificial cloud condition with a Digital Single Lens Reflex (DSLR) camera as red-green-blue (RGB) raw data throughout 2017&ndash, 2018 summer. Three supervised classifiers, including artificial neural networks (ANN), support vector machine (SVM), and k-nearest neighbors (kNN) were evaluated for disease detection. A total of 40 textural features were extracted using a colour co-occurrence matrix (CCM). The collected feature data were normalized, then used for training and internal, external and cross-validations of developed classifiers. Results of this study revealed that the highest overall classification accuracy was 93.81% using the ANN classifier and lowest overall accuracy was 78.80% using the kNN classifier. Results identified the ANN classifier disease detection having a lower Root Mean Square Error (RMSE) = 0.004 and Mean Absolute Error (MAE) = 0.003 values with 99.99% of accuracy during internal validation and 87.41%, 88.95% and 95.04% of accuracies during external validations with three different fields. Overall results demonstrated that an image texture-based ANN classifier was able to classify PM disease more accurately at early stages of disease development.

Existing texture filtering techniques acquiesce in the removal of all undesired background texture patterns simultaneously, and lead to the disappearance of small-scale foreground structures. Differently, this paper considers both multi-scale structures and specified texture patterns within an image as significant image contents, and presents a content-aware texture filtering approach via an easy-to-use interactive interface, named intuitive single click. The user’s effort on content-aware texture filtering is reduced to only locating a single click, where it is intuitively considered as one type of texture patterns. This simple interface is made possible by first generating three-scale intermediate filtered images (i.e., high, medium and low) and then performing breadth-first search (BFS) to identify one or more texture regions. Specifically, we utilize BFS to purify image edges in the medium-scale filtered image, on top of which a mask representing texture and non-texture regions is abstracted through the intuitive single-click texture locating and BFS. The mask is applied to effectively maintain multi-scale structures in the low-scale filtered image, and guides to fully smooth textures in the high-scale filtered image. Finally, a clean image is yielded by simply synthesizing the results of two mask operations in the low- and high-scale filtered images. We illustrate by extensive experiments that our approach advances the state-of-the-arts in terms of both the image cleanness and content-preserving ability. Our code is publicly available.

Compared to multispectral or panchromatic bands, fusion imagery contains both the spectral content of the former and the spatial resolution of the latter. Even though the Estimation of Scale Parameter (ESP), the ESP 2 tool, and some segmentation evaluation methods have been introduced to simplify the choice of scale parameter (SP), shape, and compactness, many challenges remain, including obtaining the natural border of plastic greenhouses (PGs) from a GaoFen-2 (GF-2) fusion imagery, accelerating the progress of follow-up texture analysis, and accurately evaluating over-segmentation and under-segmentation of PG segments in geographic object-based image analysis. Considering the features of high-resolution images, the heterogeneity of fusion imagery was compressed using texture analysis before calculating the optimal scale parameter in ESP 2 in this study. As a result, we quantified the effects of image texture analysis, including increasing averaging operator size (AOS) and decreasing greyscale quantization level (GQL) on PG segments via recognition of a proposed Over-Segmentation Index (OSI)-Under-Segmentation Index (USI)-Error Index of Total Area (ETA)-Composite Error Index (CEI) pattern. The proposed pattern can be used to reasonably evaluate the quality of PG segments obtained from GF-2 fusion imagery and its derivative images, showing that appropriate texture analysis can effectively change the heterogeneity of a fusion image for better segmentation. The optimum setup of GQL and AOS are determined by comparing CEI and visual analysis.

This paper proposes a novel distortion function for JPEG steganography, which depends on the magnitude of discrete cosine transformation (DCT) coefficients, the first- and second-order residuals, a...

Salient object detection irrespective of all textural variations in scene.Eschews all heuristics and computes different weights from the image itself.Assigns a uniform saliency value to all pixels inside the salient region.Extracts the complete contour(s) of salient object(s) from an image.Identifies whether or not an image contains any salient region at all. We propose a texture insensitive, region based image saliency detection algorithm, having excellent detection and localization properties, to obtain salient objects. We use a total variation based regularizer to suppress textures from the image and to make the method invariant to textural variations in the scene. This leads to an image that contains piecewise constant gray valued regions. This texture-free image is sparsely segmented into a small number of regions using the expectation maximization algorithm assuming a Gaussian mixture model. We compute three different saliency measures for every region using its intensity and spatial features. We adopt a relevance feedback mechanism to obtain weights for combining the three saliency measures and obtain the final saliency map. Next we input the thresholded saliency map to an image matting technique and extract the salient objects from the image with exact boundaries. Experimental comparisons with existing saliency detection algorithms demonstrate the superiority of the proposed technique.

This paper introduces a new method for clustering of documents, which have been written in a language evolving during different historical periods, with an example of the Italian language. In the first phase, the text is transformed into a string of four numerical codes, which have been derived from the energy profile of each letter, defining the height of the letters and their location in the text line. Each code represents a gray level and the text is codified as a 1-D image. In the second phase, texture features are extracted from the obtained image in order to create document feature vectors. Subsequently, a new clustering algorithm is employed on the feature vectors to discriminate documents from different historical periods of the language. Experiments are performed on a database of Italian documents given in Italian Vulgar and modern Italian. Results demonstrate that this proposed method perfectly identifies the historical periods of the language of the documents, outperforming other well-known clustering algorithms generally adopted for document categorization and other state-of-the-art text-based language models.

SIFT algorithm has good effect for most images in feature matching. But it cannot extract plenty of feature points from blurred images, and the feature points extracted from low-light images have cluster effect seriously. That is because the value of contrast threshold which was used to remove the low responsive-point is fixed. This paper proposes an adaptive contrast threshold SIFT algorithm to improve this situation. The value of contrast threshold is determined by the average value of local extreme points and the texture character of the whole image. Experimental results show that the improved SIFT algorithm can extract plenty of feature points, which are effective, from blurred images and improve the distributed uniformity of the feature points extracted from low-light images.

A sign language is a mode of communication in which the intent of the message or the message itself is conveyed through body postures or the movement of the parts of the body like head, eyebrows and cheeks etc. Every expression is distinct and has distinguishable parts of language, its grammatical content fully displayed through gestures. There are more than three hundred sign languages used in the world. The baby sign language is a method of communication between the mothers and their toddlers by means of gestures, clearly expressing their emotions and desires. In the present research work study of the existing literature has been carried out and then prepared a data set of still jpeg images for 60 odd baby signs, performed GLCM(Gray Level Co-Occurrence Matrix) based feature extraction, then performed classification of gestures using KNN and Random Forest based machine learning algorithms. A classification accuracy of 73% has been achieved on the dataset prepared.

The paper presents a new image classification technique which first extracts rotation-invariant image texture features in singular value decomposition (SVD) and discrete wavelet transform (DWT) domains. Subsequently, it exploits a support vector machine (SVM) to perform image texture classification. For convenience, it is called the SRITCSD method hereafter. First, the method applies the SVD to enhance image textures of an image. Then, it extracts the texture features in the DWT domain of the SVD version of the image. Also, the SRITCSD method employs the SVM to serve as a multiclassifier for image texture features. Meanwhile, the particle swarm optimization (PSO) algorithm is utilized to optimize the SRITCSD method, which is exploited to select a nearly optimal combination of features and a set of parameters utilized in the SVM. The experimental results demonstrate that the SRITCSD method can achieve satisfying results and outperform other existing methods under considerations here. The following structure displays the conceptual design of the SRITCSD method for image texture classification. More specifically, it depicts the structure of the training phase and the testing phase for the SRITCSD method. In the training phase, the DWT-FE component denotes the feature-extraction scheme applied for a DWT version image. The feature set, f D W T j , in Eq. (8) is computed via feeding the DWT-FE component with a DWT version image. Let I S V D j represent that image I j is enhanced via the SVD. Another feature set, f S V D , D W T j , in Eq. (11) is calculated via feeding the DWT-FE component with a DWT version of I S V D j . Also, the SVM performs as a multiclassifier with respect to a set of training patterns which are constructed using image texture features, f D W T j and f S V D , D W T j . Meanwhile, the PSO algorithm is employed to optimize the SRITCSD method, which selects the nearly optimal combination of features and a set of parameters utilized in the SVM. In the testing phase of the SRITCSD method, two feature sets, f D W T q and f S V D , D W T q , are computed for a query image I q . The classification result can be obtained via feeding the trained SVM model with f S V D , D W T q to estimate which category the image I q belongs to.Display Omitted The paper presents an image classification technique which extracts rotation-invariant image texture features in singular value decomposition (SVD) and discrete wavelet transform (DWT) domains.First, the method applies the SVD to enhance image textures of an image. Then, it extracts the texture features in the DWT domain of the SVD version of the image.Also, the SVM serves as a multiclassifier for image texture features.Meanwhile, the particle swarm optimization (PSO) algorithm is exploited to select a nearly optimal combination of features and a set of parameters utilized in the SVM.The experimental results demonstrate that the method can achieve satisfying results and outperform other existing methods.

Source camera identification enables forensic investigator to discover the probable source model that are employed to acquire the image under investigation. It is important whenever digital content is presented as a silent witness. In this paper, we present a source camera identification method via image texture features that are extracted from well selected color model and color channel. Except to distinguish source camera models from images whatever they are captured via same or different brand cameras, the main contributions of the proposed method are as follows: (1) It can distinguish imaging device individuals from images even if they are taken by using same brand and model of devices. (2) It is robust for content-preserving manipulations or geometric distortions, such as JEPG compression, adding noise, and rotation and scaling. The experimental results demonstrate that the performance of the proposed method is satisfactory. Compared with the state-of-the-art methods, the proposed method is superior in both detection accuracy and robustness.

Knee Osteoarthritis (OA) is one of the most prominent diseases in an ageing society and has affected over 10 million people in Thailand. When people suffer from OA, it is very difficult to recover. Therefore, early detection and prevention are important. The typical way to detect OA is by using X-ray imaging. This research study is focused on early detection of OA by applying image processing and classification techniques to knee X-ray imagery. The fundamental concept is to find a region of interest, use feature extraction techniques and build a classifier that can classify between OA or non-OA imageries. There are four regions of interest obtained from each image: (i) Medial Femur (MF), (ii) Lateral Femur (LF), (iii) Medial Tibia (MT), and (iv) Lateral Tibia (LT). The ten texture analysis techniques are then adopted to generate the embedded properties of the bone surface. Once the feature vector has been generated the variety of techniques of machine learning mechanisms are applied to generate the desired classifiers, which can be used to distinguish between OA and non-OA images. From the conducted experiments, a total of 131 images (68 OA cases and 63 non-OA cases) was used, the results obtained show that LF region with Local Binary Pattern descriptor produced the most appropriate classifier with an AUC value of 0.912.

Threats emanating from Internet of Things (IoT) malicious software are fast evolving and complex phenomena. Unlike conventional networks, Internet of things have unique attributes such as heterogeneity of devices, high scalability and diverse architectures that makes its malware analysis complex. In this paper, we propose an approach for analyzing and classifying IoT malware using Haralick image texture features and machine learning classifiers namely K-nearest neighbor(KNN), naive Bayes (NB) and random forest(RF). A binary file (malicious or benign) is converted to a gray scale image . The gray level co-occurence matrix (GLCM) is computed on each of the extracted image. On the basis of these GLCM parameters, five Haralick features namely angular second moment, entropy, contrast, inverse different moment and correlation are calculated. Finally, these Haralick texture features are used to perform malware classification using random forest, naive Bayes and K-nearest neighbor. Experimental results shows that our approach obtains 95% accuracy for Random Forest, 89% for naive Bayes and 80% for K-nearest neighbor classifiers . Overall, use of texture feature can realize a low computational and platform independent classification scheme for IoT malware.

Recently, as demand for high-quality video and realistic media has increased, High Efficiency Video Coding has been standardized. However, High Efficiency Video Coding requires heavy cost in terms of computational complexity to achieve high coding efficiency, which causes problems in fast coding processing and real-time processing. In particular, High Efficiency Video Coding inter-coding has heavy computational complexity, and the High Efficiency Video Coding inter prediction uses reference pictures to improve coding efficiency. The reference pictures are typically signaled in two independent lists according to the display order, to be used for forward and backward prediction. If an event occurs in the input video, such as a scene change, the inter prediction performs unnecessary computations. Therefore, the reference picture list should be reconfigured to improve the inter prediction performance and reduce computational complexity. To address this problem, this article proposes a method to reduce computational complexity for fast High Efficiency Video Coding encoding using information such as scene changes obtained from the input video through preprocessing. Furthermore, reference picture lists are reconstructed by sorting the reference pictures by similarity to the current coded picture using Angular Second Moment, Contrast, Entropy, and Correlation, which are image texture parameters from the input video. Simulations are used to show that both the encoding time and coding efficiency could be improved simultaneously by applying the proposed algorithms.

This article proposes an algorithm for script identification by textural analysis of the image corresponding to the script types. In the first phase, each letter is modeled by the equivalent script type, which is determined by its position in the baseline area. Then, feature extraction is carried out. It is based on the script type cooccurrence pattern analysis. The obtained features of the script are stored for further analysis. The difference in script characteristics contributes to the diversity of the extracted features, which simplify the feature classification obtained by an extension of a state-of-the-art classification tool called Genetic Algorithms Image Clustering for Document Analysis. Accordingly, it represents the key element in the decision-making process of script identification. The proposed method is tested on an example of German printed documents, which contain Latin and Fraktur scripts. The experiment shows correct results, which is promising.

Optimal texture feature extraction in multi-class texture classification is a challenging task. The choice of traditional texture features for texture classification and segmentation is subjective and highly application dependent with lower generalization to other textures. The Deep Neural Network (DNN) approach can overcome the problem of selecting a suitable and optimal subset of texture feature extraction method. The hidden layers inside the DNN architecture automatically extract suitable features without user intervention to give the best texture discriminating performances in multi-class texture classification environment. Our results on publicly available texture datasets show that deep learning techniques could be successfully used in image texture classification and segmentation.

In this paper, we focused on building a universal haptic texture models library and automatic assignment of haptic texture models to any given surface from the library based on image features. It is shown that a relationship exists between perceived haptic texture and its image features, and this relationship is effectively used for automatic haptic texture model assignment. An image feature space and a perceptual haptic texture space are defined, and the correlation between the two spaces is found. A haptic texture library was built, using 84 real life textured surfaces, by training a multi-class support vector machine with radial basis function kernel. The perceptual space was classified into perceptually similar clusters using K-means. Haptic texture models were assigned to new surfaces in a two step process; classification into a perceptually similar group using the trained multi-class support vector machine, and finding a unique match from within the group using binarized statistical image features. The system was evaluated using 21 new real life texture surfaces and an accuracy of 71.4 percent was achieved in assigning haptic models to these surfaces.

International audience; Complexity measures, defined as measures of irregularity over time scales, are the subject of a growing number of studies as the information they reveal can find utility in a large field of applications. One of the most popular complexity measures is the multiscale entropy. Nevertheless, more and more algorithms dedicated to complexity analyses are proposed to improve the existing ones. However, such measures are available only for one-dimensional time series. For bidimensional data(images), no equivalent algorithm has been proposed to analyze irregularity over spatial scales. We herein introduce a new framework that extends the one-dimensional multiscale entropy(MSE1D) to the bidimensional case(MSE2D). Moreover, a variant of MSE2D is also ModMSE2D). The two new algorithms are tested as new texture analysis frameworks. They are applied to simulated and real data. Our results show that, compared with other existing texture analysis algorithms, MSE2D and ModMSE2D are suitable and powerful tools for image analysis and classification according to their texture patterns. While MSE2D is computationally faster than ModMSE2D, ModMSE2D is more robust to small image sizes. The two methods present interesting performances, and can be as useful as their unidimensional versions in two-dimensional applications.

Texture classification is a problem widely studied in computer vision, there exist two fundamental issues: how to describe texture images and how to define a similarity measure. The texture descriptors are mainly used to extract and represent the features of texture images and their performance is usually measured using a classification algorithm. In this paper, some of the most referenced texture descriptors, such as Gabor filter banks, Wavelets, and Local Binary Patterns, are compared using non-parametric statistical tests to know if there is a difference in performance. The descriptors are applied to five well-known texture image datasets, in order to be classified. Three classification algorithms, with a cross-validation scheme, are used to classify the described texture datasets. Finally, a Friedman test with multiple comparisons is used to compare the whole performance of the texture descriptors on a statistical basis. The statistical results suggest that for these tests there is a difference in performance, so it was possible to determine statistically, for the considered experimental settings, the best texture descriptor.

A static gesture recognition algorithm is proposed based on a recursive graph of the upper triangular image texture, motivated by the low accuracy and robustness of existing algorithms. Firstly, the fingertip localization method based on contour curvature is used to obtain the palm region and then the gesture contour model is established. Secondly, a recurrence plot of the gesture contour sequence is built, which is constructed using the central point and the starting point coordinates. Finally, the texture recognition algorithm is applied to calculate the normalized distance between the recurrence plots of the gesture. The experimental results show that the proposed algorithm can achieve higher recognition accuracy under varying complex backgrounds and illumination. At the same time, when the gesture is in rotation, translation, or scaling, the algorithm has high robustness with a small amount of computation and high efficiency.

This paper introduces an image texture analysis method for minority language identification. In the first stage, each letter is associated with a given script type according to its energy status in the text-line area. Mapping is carried out by extracting unicode text and transforming it into coded text. There are four different script types, which correspond to four grey levels of an image. Then, the obtained image is subjected to a feature extraction process performed by the texture analysis. This way, the grey level co-occurrence matrix and its derivative features are calculated. Extracted features are compared and classified using the K-Nearest Neighbors and Naive Bayes methods to establish a difference that can identify a minority language such as Serbian language among other world languages in the text. Very good accuracy results prove the efficiency of the proposed approach, when compared to other state-of-the-art methods.

We present an algorithm for risk prediction of road surface grip where skidding and sliding occur as main road surface problems. Prediction is done by defining a fine texture classification of the properties of road aggregate. In an experimental setup, data acquisition is performed with a supervised mobile vehicle scanning system, using a vehicle equipped with a camera and temperature sensor during movement along an arterial road. Image processing is performed by testing four texture feature extraction methods: Gabor filters, wavelet transform, gray level co-occurence matrix, and edge histogram descriptor, among which the Gabor transform shows the best results. The extraction of texture feature vectors follows by statistical algorithms for measuring feature vector similarity and reference vector selection, leading to image texture classification. The algorithm itself is upgraded by incorporating simultaneous surface temperature measurements in order to create and validate the final fine surface texture classification. The roads are classified and segmented into high-, medium-, and low-risk roads according to skid danger, enabling the creation of a map of high-risk zones. We validate our risk prediction algorithm by referring to crash rate data from the Road Traffic Safety Agency of Serbia database. This algorithm enables the location and mapping of high-risk zones and can be used as a support for autonomous driving and navigation.

Public transport vehicle scheduling system plays a very important role in relieving urban traffic pressure, and it's now still a difficult problem. In order to solve the problem effectively, we propose a bus scheduling method based on video processing algorithm in this paper. Firstly, we decode the H264 bare stream which receive from the bus in real-time, and according to the prior knowledge, a certain number of training samples with different density levels were collected, and then extract the improved gray-level co-occurrence matrix and color entropy features of training samples to train the classifier model. Finally, crowd density estimation result can get by processing the real time vehicle video using the trained model. Combined the result of density estimation with vehicle status information, we can realize the vehicle scheduling effectively.

We introduce a microstructure informatics dataset focusing on complex, hierarchical structures found in a single Ultrahigh carbon steel under a range of heat treatments. Applying image representations from contemporary computer vision research to these microstructures, we discuss how both supervised and unsupervised machine learning techniques can be used to yield insight into microstructural trends and their relationship to processing conditions. We evaluate and compare keypoint-based and convolutional neural network representations by classifying microstructures according to their primary microconstituent, and by classifying a subset of the microstructures according to the annealing conditions that generated them. Using t-SNE, a nonlinear dimensionality reduction and visualization technique, we demonstrate graphical methods of exploring microstructure and processing datasets, and for understanding and interpreting high-dimensional microstructure representations., Comment: Data publication forthcoming

Blind noise-level estimation (NLE) is a fundamental issue in digital image processing. This paper provides a noise-level estimator for additive white Gaussian noise and multiplicative Gaussian noise using principal texture patches (PTPs). The proposed algorithm first identifies the principal texture of the noisy image by using the principal component analysis, and then, it chooses PTPs to calculate the noise level. The major contributions of this paper toward addressing the challenges in the NLE literature include: 1) analyzing the nonlinear relationship between the smallest eigenvalue, true noise level, number of chosen patches, and block size; 2) extracting PTPs; and 3) computing noise level using PTPs. The abundant experimental results show that the proposed method works well for various natural images over a large range of noise levels and it performs well for multiplicative noise. Compared with some state-of-the-art approaches, our method has the best performance and a faster running speed.

In the present paper, a methodological scheme, bringing together common Acoustic Seabed Classification (ASC) systems and a powerful data decomposition approach, called Independent Component Analysis (ICA), is demonstrated regarding its suitability for detecting small targets in Side Scan Sonar imagery. Traditional ASC systems extract numerous texture descriptors, leading to a large feature vector, the dimensionality of which is reduced by means of data decomposition techniques, usually Principal Component Analysis (PCA), prior to classification. However, in the target detection issue, data decomposition should point towards finding components that represent sub-ordinary image information (i.e., small targets) rather than a dominant one. ICA has long been proved to be suitable for separating targets from a background, and this study represents a novel exhibition of its applicability to Side Scan Sonar (SSS) images. The present study attempts to build a fully automated target detection approach that combines image based feature extraction, ICA, and unsupervised classification. The suitability of the proposed approach has been demonstrated using an SSS data-set containing more than 70 manmade targets, most of them metallic, validated through a marine magnetic survey or ground truthing inspection. The method exhibited very good performance as it was able to detect more than 77% of the targets and it produced less than seven false alarms per km2. Moreover, it was compared to cases where, in the exact same methodological scheme, no decomposition technique is used, or PCA is employed instead of ICA, achieving the highest detection rate, but, more importantly, producing more than six times less false alarms, thus proving that ICA successfully manages to maximize target to background separation.

Semi-variogram estimators and distortion measures of signal spectra are utilized in this paper for image texture retrieval. On the use of the complete Brodatz database, most high retrieval rates are reportedly based on multiple features and the combinations of multiple algorithms, while the classification using single features is still a challenge to the retrieval of diverse texture images. The semi-variogram, which is theoretically sound and the cornerstone of spatial statistics, has the characteristics shared between true randomness and complete determinism and, therefore, can be used as a useful tool for both the structural and statistical analysis of texture images. Meanwhile, spectral distortion measures derived from the theory of linear predictive coding provide a rigorously mathematical model for signal-based similarity matching and have been proven useful for many practical pattern classification systems. Experimental results obtained from testing the proposed approach using the complete Brodatz database, and the the University of Illinois at Urbana-Champaign texture database suggests the effectiveness of the proposed approach as a single-feature-based dissimilarity measure for real-time texture retrieval.

International audience; Texture feature is a measure method about relationship among the pixels in local area, reflecting the changes of image space gray levels. This paper presents a texture feature extraction method based on regional average binary gray level difference co-occurrence matrix, which combined the texture structural analysis method with statistical method. Firstly, we calculate the average binary gray level difference of eight-neighbors of a pixel to get the average binary gray level difference image which expresses the variation pattern of the regional gray levels. Secondly, the regional co-occurrence matrix is constructed by using these average binary gray level differences. Finally, we extract the second-order statistic parameters reflecting the image texture feature from the regional co-occurrence matrix. Theoretical analysis and experimental results show that the image texture feature extraction method has certain accuracy and validity

The paper proposes an algorithm for the script recognition based on the texture characteristics. The image texture is achieved by coding each letter with the equivalent script type (number code) according to its position in the text line. Each code is transformed into equivalent gray level pixel creating an 1-D image. Then, the image texture is subjected to the run-length analysis. This analysis extracts the run-length features, which are classified to make a distinction between the scripts under consideration. In the experiment, a custom oriented database is subject to the proposed algorithm. The database consists of some text documents written in Cyrillic, Latin and Glagolitic scripts. Furthermore, it is divided into training and test parts. The results of the experiment show that 3 out of 5 run-length features can be used for effective differentiation between the analyzed South Slavic scripts., 9 pages, 9 figures, In Electronics 2015, Elektronika IR Elektrotechnika, ISSN 1392-1215 (in press)

This chapter presents variational models to perform texture analysis, extraction, or both for image processing. It focuses on second order decomposition models. Variational decomposition models have been studied extensively during the past decades. The most famous one is the Rudin-Osher-Fatemi model. First, most classical first order, models are discussed. Then the chapter deals with second order ones: the mathematical framework, theoretical models, and numerical implementation, ending with two 3D applications. Finally, an appendix includes the mathematical tools that are used to perfom this study and a second appendix provides Matlab© codes, ending.

The Kolmogorov-Sinai (K-S) entropy is used to quantify the average amount of uncertainty of a dynamical system through a sequence of observations. Sequence probabilities therefore play a central role for the computation of the entropy rate to determine if the dynamical system under study is deterministically non-chaotic, deterministically chaotic, or random. This paper extends the notion of the K-S entropy to measure the entropy rate of imprecise systems using sequence membership grades, in which the underlying deterministic paradigm is replaced with the degree of fuzziness. While constructing sequential probabilities for the calculation of the K-S entropy is difficult in practice, the estimate of the K-S entropy in the setting of fuzzy sets in an image is feasible and can be useful for modeling uncertainty of pixel distributions in images. The fuzzy K-S entropy is illustrated as an effective feature for image analysis and texture classification. HighlightsDevelopment of a new method for feature extraction using chaos and fuzzy sets.Modeling Kolmogorov-Sinai (K-S) entropy with fuzzy logic is feasible.Fuzzy K-S entropy is an effective feature for analysis and classification of textural images.

Image texture analysis is a key task in computer vision. Although various methods have been applied to extract texture information, none of them are based on the principles of sample entropy, which is a measurement of entropy rate. This paper proposes a two-dimensional sample entropy method, namely SampEn2D, in order to measure irregularity in pixel patterns. We evaluated the proposed method in three different situations: a set of simulated images generated by a deterministic function corrupted with different levels of a stochastic influence; the Brodatz public texture database; and a real biological image set of rat sural nerve. Evaluation with simulations showed SampEn2D as a robust irregularity measure, closely following sample entropy properties. Results with Brodatz dataset testified superiority of SampEn2D to separate different image categories compared to conventional Haralick and wavelet descriptors. SampEn2D was also capable of discriminating rat sural nerve images by age groups with high accuracy (AUROC = 0.844). No significant difference was found between SampEn2D AUROC and those obtained with the best performed Haralick descriptors, i.e. entropy (AUROC = 0.828), uniformity (AUROC = 0.833), homogeneity (AUROC = 0.938) and Wavelet descriptors, i.e. Haar energy/entropy (AUROC = 0.932) and Daubechies energy/entropy (AUROC = 0.859). In addition, it was shown that SampEn2D computation time increases with image size, being around 1400 s for a 600 × 600 pixels image. In conclusion, SampEn2D showed to be stable and robust enough to be applied as texture feature quantifier and irregularity properties, as measured by SampEn2D, seem to be an important feature for image characterization in biomedical image analysis.

This paper investigates texture feature capabilities from fundus images to differentiate between diabetic retinopathy (DR), age-related macular degeneration (AMD) screening and normal. Our proposed method using improvement of local binary pattern (LBP) with calculation of LBP original value and magnitude value of fundus images. This method is compared with Local Line Binary Pattern (LLBP). In this study, four experiments (DR-Normal, DR-AMD, AMD-Normal, Multiclass) were designed for two databases, DIARETDB0 database and STARE. Kernel PCA is choosed as feature selection method, and three classifiers are tested (Naive Bayes, SVM, and KNN). The experimental results show that our proposed method has higher accuracy than LLBP, with accuracy of binary classification 100% for DR-Normal and AMD-Normal. While, multiclass classification (DR-AMD-Normal) achieves an accuracy 80–84%. These results suggest that our proposed method in this paper can be useful in a diagnosis aid system for diabetic retinopathy.

The existing methods—such as sieving, microscope, light scattering, sedimentation, and electrical induction for pigment size detection—require sampling or scattering the mineral pigments, which will inevitably cause damage to the films painted by mineral pigments. A new detection method based on run length texture analysis is proposed to nondestructively detect the pigment size in the mineral paint film. The films painted by mineral pigments with preknown pigment sizes are contactlessly captured by CCD microscope under diffused light. Gray transform, histogram equalization, and median filtering are implemented to preprocess the captured images, and then the run length texture parameters are extracted from the preprocessed images. A parametric relationship between the extracted parameters and the preknown size is established to predict the pigment size in mineral paint film nondestructively. Burnt carnelian is selected as the sample to verify the feasibility of the proposed method. Results show that the max detection error of the proposed method is 5.548 μm and can be applied to the size detection of the mineral pigments used in mineral paint film.

This study investigated both spectrum and texture features for detecting early blight disease on eggplant leaves. Hyperspectral images for healthy and diseased samples were acquired covering the wavelengths from 380 to 1023 nm. Four gray images were identified according to the effective wavelengths (408, 535, 624 and 703 nm). Hyperspectral images were then converted into RGB, HSV and HLS images. Finally, eight texture features (mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment and correlation) based on gray level co-occurrence matrix (GLCM) were extracted from gray images, RGB, HSV and HLS images, respectively. The dependent variables for healthy and diseased samples were set as 0 and 1. K-Nearest Neighbor (KNN) and AdaBoost classification models were established for detecting healthy and infected samples. All models obtained good results with the classification rates (CRs) over 88.46% in the testing sets. The results demonstrated that spectrum and texture features were effective for early blight disease detection on eggplant leaves.

We analyze applicability of 2D fractal Brownian motion (fBm) for real-life image textures with respect to two general fBm properties: isotropy and normality of its increments. A non-parametric detection scheme for texture satisfying these two properties is proposed. It is based on Lilliefors test for texture increments normality and Kolmogorov-Smirnov two samples test for equality of distributions of pairs of increments. The scheme is tested against large real-life images database and is shown to detect and remove such image patterns as edges, areas with clipping effects, irregular and anisotropic textures.

This paper deals with the development of a new texture analysis method based on both spatial and spectral information for texture classification purposes. The idea of the Spatial and Spectral Gray Level Dependence Method (SSGLDM) is to extend the concept of spatial gray level dependence method by assuming texture joint information between spectral bands. In addition, new texture features measurement related to (SSGLDM) which define the image properties have been also proposed. Extensive experiments have been carried out on many multi-spectral images for use in prostate cancer diagnosis and quantitative results showed the efficiency of this method compared to the Gray Level Co-occurrence Matrix (GLCM). The results indicate a significant improvement in classification accuracy.

We investigate a model for textured images using a functional minimization problem. We decompose a textured image ud as a sum of two functions u+v, where u is a bounded variation function and v is a “ bounded hessian” function. We also present numerical tests.

The internet is a big giant in the today’s world playing the backbone for information and communication technologies. Most of the network application needs a very reliable and one-way authentication process to ensure system resources are securely accessed by authorised users over the Internet. Password-based authentication is the most common mechanism and is easy to implement for authentication purpose. However, such general scheme will be vulnerable to attackers to steal the static credential combination with ease using eavesdropping, brute forcing, password replay, etc. One-time password (OTP) is the promising solutions to overcome from such attacks. It generates the different unique passwords each time for the user login into the system. This paper proposes a novel method of OTP authentication using image features. Here the system calculates different features from randomly selected areas of an image and applies symmetric key cryptography to generate the random OTP for the user. The proposed approach is convenient and computationally less expensive offering high-level security.