Search

The purpose of image smoothing is to improve the visual effect of the image and improve the clarity of the image, so as to make the image more conducive to computer processing and various feature analysis. Because the current technology fails to smooth the preprocessed image, it leads to the extraction of image texture features. The anti-interference performance is weak. For this reason, an image texture feature extraction technology based on the digital twin is proposed. Similarity analysis is carried out through the internal structure of the image, and the image is smoothed by the semisupervised learning method. On the basis of optimizing the denoised image through digital twinning, detect target feature points in the original image, then remove the abnormal and split feature points, assign the direction of image texture feature points, and build a fuzzy back propagation neural network model. Image texture feature extraction technology is implemented. The experimental results show that, compared with the traditional method, the proposed technique has a strong identification of original image features, and has a strong consistency with original data, and has a strong ability to resist the influence of abnormal data, noise, or redundant feature points. [ABSTRACT FROM AUTHOR]

Up-to-date information regarding impervious surface is valuable for urban planning and management. The objective of this study is to develop neural computing models used for automatic impervious surface area detection at a regional scale. To achieve this task, advanced optimizers of adaptive moment estimation (Adam), a variation of Adam called Adamax, Nesterov-accelerated adaptive moment estimation (Nadam), Adam with decoupled weight decay (AdamW), and a new exponential moving average variant (AMSGrad) are used to train the artificial neural network models employed for impervious surface detection. These advanced optimizers are benchmarked with the conventional gradient descent with momentum (GDM). Remotely sensed images collected from Sentinel-2 satellite for the study area of Da Nang city (Vietnam) are used to construct and verify the proposed approach. Moreover, texture descriptors including statistical measurements of color channels and binary gradient contour are employed to extract useful features for the neural computing model-based pattern recognition. Experimental result supported by statistical test points out that the Nadam optimizer-based neural computing model has achieved the most desired predictive accuracy for the data collected in the studied region with classification accuracy rate of 97.331%, precision = 0.961, recall = 0.984, negative predictive value = 0.985, and F1 score = 0.972. Therefore, the model developed in this study can be a helpful tool for decision-makers in the task of urban land-use planning and management. [ABSTRACT FROM AUTHOR]

A new algorithm based on wavelet packet transform and singular value decomposition is proposed in this paper for classification of textures in the presence of noise. Lower singular values are affected more by noise than higher singular values, and hence only higher singular values are used to classify textures in the presence of noise. The probability density function of the selected singular values is then modeled as an exponential distribution, and the model parameter for the distribution is estimated using the maximum likelihood estimation technique. The model parameter, one for each subband is used as features for the classification. The classification is carried out using Weighted Probabilistic Neural Networks (WPNN). Compared to conventional probabilistic neural networks, WPNN includes weighting factors between pattern layer and summation layer of the conventional PNN. Performance of the algorithm is compared with `wavelet domain generalized Gaussian density based model' in terms of signal to noise ratio and classification rate. Experimental results prove that the proposed algorithm achieves better classification rate under noisy environment. [ABSTRACT FROM AUTHOR]

The suitability of optical IKONOS satellite data (multispectral and panchromatic) for the estimation of forest structural attributes – for example, stems per hectare (SPHA), diameter at breast height (DBH), mean tree height (MTH), basal area (BA) and volume in plantation forest environments – was assessed in this study. The relationships of these forest structural attributes to statistical image texture from IKONOS imagery were analysed. The coefficients of determination (R 2) of multilinear regression models developed for the estimation of SPHA, DBH, MTH, BA and volume using statistical texture features from multispectral data were 0.63, 0.68, 0.81, 0.86 and 0.86, respectively. When the statistical texture features from panchromatic data were applied, the R 2 for the respective forest structural attributes increased by 25%, 31%, 6%, 0.2% and 0.2%, respectively. Artificial neural network (ANN) models produced strong and significant relationships between estimated and actual measures of SPHA, DBH, MTH, BA and volume with an R 2 of 0.83, 0.83, 0.90, 0.90 and 0.92, respectively, based on multispectral IKONOS data. Based on panchromatic IKONOS imagery, the R 2 for the respective forest structural attributes increased by 18%, 12%, 5%, 3% and 6%, respectively. Results such as these bode well for the application of high spatial resolution imagery to forest structural assessment. [ABSTRACT FROM AUTHOR]

Abstract: Quantifying the above ground biomass of tropical forests is critical for understanding the dynamics of carbon fluxes between terrestrial ecosystems and the atmosphere, as well as monitoring ecosystem responses to environmental change. Remote sensing remains an attractive tool for estimating tropical forest biomass but relationships and methods used at one site have not always proved applicable to other locations. This lack of a widely applicable general relationship limits the operational use of remote sensing as a method for biomass estimation, particularly in high biomass ecosystems. Here, multispectral Landsat TM and JERS-1 SAR data were used together to estimate tropical forest biomass at three separate geographical locations: Brazil, Malaysia and Thailand. Texture measures were derived from the JERS-1 SAR data using both wavelet analysis and Grey Level Co-occurrence Matrix methods, and coupled with multispectral data to provide inputs to artificial neural networks that were trained under four different training scenarios and validated using biomass measured from 144 field plots. When trained and tested with data collected from the same location, the addition of SAR texture to multispectral data showed strong correlations with above ground biomass (r =0.79, 0.79 and 0.84 for Thailand, Malaysia and Brazil respectively). Also, when networks were trained and tested with data from all three sites, the strength of correlation (r =0.55) was stronger than previously reported results from the same sites that used multispectral data only. Uncertainty in estimating AGB from different allometric equations was also tested but found to have little effect on the strength of the relationships observed. The results suggest that the inclusion of SAR texture with multispectral data can go someway towards providing relationships that are transferable across time and space, but that further work is required if satellite remote sensing is to provide robust and reliable methodologies for initiatives such as Reducing Emissions from Deforestation and Degradation (REDD+). [Copyright &y& Elsevier]

Deep neural networks have been extensively applied in fields such as image classification, object detection, and face recognition. However, research has shown that adversarial samples with subtle perturbations can effectively deceive these networks. Existing methods for generating such adversarial images often lack stealth and robustness. In this study, we present an enhanced attack strategy based on traditional Generative Adversarial Networks (GANs). We integrate image texture into the unsupervised training scheme, guiding the model to focus perturbations in high-texture areas. We also introduce a dynamic equilibrium training strategy that employs Differential Evolution algorithms to adaptively adjust both network weight parameters and the training ratio between the generator and discriminator, achieving a self-balancing training process. Further, we propose an image local optimization algorithm to eliminate perturbations in non-sensitive areas through weighted filtering. The model is validated using benchmark datasets such as MNIST, ImageNet and SVHN. Through extensive experimental evaluations, our approach shows a 4.93% improvement in attack success rate against conventional models and a 10.23% increase against defense models compared to state-of-the-art attack methods. [ABSTRACT FROM AUTHOR]

Biospeckle activity (during 1-25 s) of ‘Red Delicious’ apples was recorded under 680 and 780 nm laser diodes. The time history speckle pattern (THSP) images were configured and evaluated using the common biospeckle and texture descriptors. Several artificial neural networks (ANN) models were developed to predict the stiffness and juiciness of apple fruits. At the recording time of 1 s and 680 nm laser diode, the ANN models developed by texture features showed better performance in prediction of stiffness (r = 0.70 and SEP = 8.4 kN) and juiciness (r = 0.68 and SEP = 2.3 cm2) as compared to the common biospeckle features-based ones (r = 0.25 and SEP = 11.3 kN m−1 for stiffness and r = 0.29 and SEP = 3.1 cm2for juiciness). Furthermore, classification of apples into fresh, semi-mealy, and mealy groups was carried out. At the recording time of 25 s and 680 nm laser, the ANN models developed by texture features classified fresh, semi-mealy, and mealy apples with accuracies of 86.3, 65.9, and 95.5%, respectively. Reduction of the recording time to 1 s resulted in the classification accuracies of 88.2, 68.5, and 85.5% for mealy, semi-mealy, and mealy apples compared to the common biospeckle features-based ANN models with the accuracies of 82.8, 50.7, and 46.8%, respectively. These results indicated that the texture descriptors had better performance at shorter recording times in comparison with the biospeckle features which are commonly used in evaluation of biospeckle images, even though more improvements are still required. [ABSTRACT FROM AUTHOR]

The vision-based grasp detection method is an important technique used to research the grasping task of robots. Unfortunately, the performances of these research methods in practice are worse than the state-of-the-art accuracy on public datasets, because shifts in data distribution owing to real-world conditions are common in the real world and neural network-based methods are sensitive to small data changes. These disturbances mainly change the image texture, causing the performance of grasp detection methods to decline sharply. However, the evaluation metric of existing models does not reflect the actual robustness of the method. Therefore, we propose a new solution. First, referring to the existing research on image classification methods, we propose a benchmark to verify the realistic robustness of the grasp detection model. Second, to improve the robustness of the model, we randomly transfer texture knowledge from other images to provide variable texture information for network training. This forces the model to rely more on the contour features of the object than on the texture when making decisions; we call this approach “context-free.” We verify the effectiveness of our method for robustness enhancement in various grasp tasks and test the proposed method in a real robot grasping scene. [ABSTRACT FROM AUTHOR]

This study was aimed at assessing the effect of microwave drying at 100, 200, or 300 W on the quality of Cavendish banana slices without pretreatment and with pretreatment using 5% ascorbic acid solution, 5% citric acid solution, 5% gum arabic solution, and ultrasound. Banana slices were imaged using a digital single-lens reflex (SLR) camera. The acquired images were processed to extract texture parameters. The classification models were developed based on image texture parameters selected from a big dataset of 2172 textures of images in different color channels using artificial neural networks. Wide Neural Network, Bilayered Neural Network, Medium Neural Network and three classifiers from the group of function, such as RBF (Radial basis function) Network, Multilayer Perceptron, and WiSARD were applied. Banana slices belonging to 15 classes with different combinations of pretreatment and microwave drying were distinguished with an average accuracy of up to 97.2% for a model built using Multilayer Perceptron. For most models, banana samples microwave-dried at 200 W without pretreatment were classified with the highest correctness. The performed study revealed that the objective, non-destructive, correct, and robust quality assessment of pretreated and microwave-dried banana slices may be performed using image processing and artificial intelligence. • Banana slices were subjected to different combinations of pretreatment and microwave drying. • Samples were classified using image processing and artificial neural networks. • The obtained classification accuracy reached 97.2%. • The most successful model was developed using Multilayer Perceptron. [ABSTRACT FROM AUTHOR]

In this study, the effects of drying medium temperature and velocity were surveyed on the image texture features of shrimp (Penaeusspp.) batches in a dryer equipped with a perpendicular dual-view computer vision system (CVS). This was carried out by applying an innovative rotation- and scale-invariant image texture processing approach with the capability of eliminating the effects of sample shrinkage on the visual textural features. Moreover, the variations in image texture parameters were investigated with moisture ratio, color, and geometrical characteristics of the shrimp samples. Drying experiments were conducted at hot air drying (HAD) temperatures of 50–90°C and superheated steam drying (SSD) temperatures of 110–120°C with drying medium velocities of 1–2 m/s. Several configurations of a multilayer perceptron artificial neural network (MLP-ANN) were also used to predict the moisture ratio and the geometrical characteristics of the shrimp batch using the image texture parameters. Generally, the image texture features were significantly affected by drying medium temperatures (p < 0.01), and the effects of drying medium velocities on the textural properties were nonsignificant (p > 0.05). Additionally, the higher drying temperatures generated products with uniform and regular texture patterns. The SSD produced samples with somewhat nonuniform and irregular texture patterns compared with HAD at 90°C. Finally, selected MLP-ANN topologies successfully predicted the moisture ratio and the geometrical characteristics of the shrimp batch using the textural properties with correlation coefficients higher than 0.99. [ABSTRACT FROM AUTHOR]

Nowadays, the information processing capabilities and resource storage capabilities of computers have been greatly improved, which also provides support for the neural network technology. Convolutional neural networks have good characterization capabilities in computer vision tasks, such as image recognition technology. Aiming at the problem of high similarity image recognition and classification in a specific field, this paper proposes a high similarity image recognition and classification algorithm fused with convolutional neural networks. First, we extract the image texture features, train different types, and different resolution image sets and determine the optimal texture different parameter values. Second, we decompose the image into subimages according to the texture difference, extract the energy features of each subimage, and perform classification. Then, the input image feature vector is converted into a one-dimensional vector through the alternating 5-layer convolution and 3-layer pooling of convolutional neural networks. On this basis, different sizes of convolution kernels are used to extract different convolutions of the image features, and then use convolution to achieve the feature fusion of different dimensional convolutions. Finally, through the increase in the number of training and the increase in the amount of data, the network parameters are continuously optimized to improve the classification accuracy in the training set and in the test set. The actual accuracy of the weights is verified, and the convolutional neural network model with the highest classification accuracy is obtained. In the experiment, two image data sets of gems and apples are selected as the experimental data to classify and identify gems and determine the origin of apples. The experimental results show that the average identification accuracy of the algorithm is more than 90%. [ABSTRACT FROM AUTHOR]

Algorithms for classifying cloud images based on MODIS and VIIRS satellite data using artificial neural networks and fuzzy logic methods are considered. A combined classification of recognized cloud types is presented. The results of using the cloud classification for solving some problems of climatology and meteorology are analyzed. A statistical model of the image texture for various cloud types and physical parameters of clouds based on two-parameter distributions is proposed. A description of the algorithm for detecting weather fronts and determining their types from satellite data and the results of its testing for the territory of Western Siberia are presented. An approach to studying internal waves in the atmosphere based on the analysis of their cloud manifestation parameters is considered. The results are presented of studying the long-term variability for some of their parameters over the water area of the Kuril Islands. The methodology and results are discussed of studying the long-term variability of the structure of global cloud fields and their parameters over the natural zones of Western Siberia in summer during 2001 to 2019. [ABSTRACT FROM AUTHOR]

In this study, the seeds of open-pollinated winter rapeseed cultivars, hybrid winter rapeseed cultivars, open-pollinated spring rapeseed cultivars and hybrid spring rapeseed cultivars were investigated. The physical, optical, mechanical, geometric and image texture properties of rapeseeds were compared. Statistical models were developed based on the analyzed parameters to discriminate between seed groups. Most parameters effectively discriminated between cultivars of winter and spring rapeseed, including true density, porosity, L* , a* , b* , and spectral values at 400 nm, 470 nm, 500–530 nm, 560–620 nm, 640–650 nm and 690 nm. Four homogeneous groups were identified based on linear dimensions: F (surface area), S (width) and shape factors W 6 (circularity ratio), Rb (Blair-Bliss coefficient) and W 13 (roundness). No statistically significant differences in the mean values of hardness or area under the force-displacement graph were observed between seed groups. The model developed based on image texture variables from channel Y (luminance) was characterized by the highest discrimination accuracy of 82–87%. The experimental groups were classified with 89–92% accuracy in the model combining the best variables from each group of physical parameters. Total classification accuracy in neural networks reached 75% for a validation set comprising geometric properties and 91–92% for a validation set containing physical characteristics. [ABSTRACT FROM AUTHOR]

Generative Adversarial Networks (GANs) emerged thanks to the development of deep neural networks. Forgery images generated by various variants of GANs are widely spread on the Internet, which may be damage personal credibility and cause huge property losses. Thus, numerous methods are proposed to detect forgery images, but most of them are designed to detect forgery faces. Therefore, a method to detect forgery images of various scenes is proposed. In this work, central difference convolution and vanilla convolution (CDC‐Mix) are mixed after considering the depth and width features of neural networks and analyzing the influence of attention on network performance. Based on CDC‐Mix, a separable convolution (SeparableCDC‐Mix) is proposed. The proposed method consists of three parts: (1) CDC‐Mix and SeparableCDC‐Mix are used to extract the gradient information and texture features; (2) CDCM is used to extract the multi‐scale information of the image; (3) multi‐scale fusion module (MS‐Fusion) is used to fuse the multi‐scale information from different locations of the network. A large number of experiments have been carried out on several datasets generated by GAN, and the experimental results show that the proposed method has a great improvement compared with the existing advanced methods. [ABSTRACT FROM AUTHOR]

[Objectives] The main goal of the article is to quickly obtain the chlorophyll content in plants and improve the accuracy of non-destructive measurement of chlorophyll. [Methods] Taking the leaves of Epipremnum aureum as the research object, a non-destructive detection method of chlorophyll SPAD value combining hyperspectral and texture features in series was proposed, the spectral information of 320 leaf samples of E.aureum in the wavelength range(400-900 nm) was collected, original hyperspectral image was preprocessed by Savitzky-Golay convolution smoothing, and 10 characteristic wavelengths were selected through successive projections algorithm(SPA);At the same time, gray-level co-occurrence matrix(GLCM) was used to extract texture features of RGB images in hyperspectral images; Finally, the concatenation method was used to fuse the hyperspectral feature and the texture feature to obtain the fusion feature. The back propagation artificial neural network(BPANN) and the support vector machine regression(SVR) models of single feature and fusion feature were established respectively. [Results] The test results showed that the prediction model established by using feature spectrum data or image texture data as feature values alone had unstable comprehensive performance and that the accuracy of the prediction model based on series fusion features had been significantly improved. The SVR model established after fusing features had the best prediction result, the correction set decision coefficient R² was 0.961 2,and the prediction set decision coefficient R² was 0.957 1. [Conclusions] This study proves that the fusion of hyperspectral features and texture features can improve the accuracy of the chlorophyll regression prediction model, and provides an important reference for the non-destructive detection of chlorophyll content. [ABSTRACT FROM AUTHOR]

This paper presents a deep learning-based image texture recognition system. The methodology taken in this solution is formed in a bottom-up manner. It means we swipe a moving window through the image in order to categorize if a given region belongs to one of the classes seen in the training process. This categorization is done based on the Deep Neural Network (DNN) of fixed architecture. The training process is fully automated regarding the training data preparation, investigation of the best training algorithm, and its hyper-parameters. The only human input to the system is the definition of the categories for further recognition and generation of the samples (region markings) in the external application chosen by the user. The system is tested on road surface images where its task is to categorize image regions to a different road category (e.g. curb, road surface damage, etc.) and is featured with 90% and above accuracy. [ABSTRACT FROM AUTHOR]

Stereo dense image matching normally refers to per-pixel correspondence search between stereo pairs. However, their matching results actually depend on matching window sizes. Large windows usually obtain robust matching results in weak-textured regions, but serious mismatches in depth/disparity (parallax in the epipolar space) inconsistency regions. Small windows compute accurate matching results in depth/disparity inconsistency regions, but it may contain high matching uncertainties in weak-textured regions. To improve matching accuracies, this letter focuses on adaptively selecting appropriate matching windows for each pixel. In general, we propose a window size selection network (WSSN) with the basic assumption that appropriate window sizes are related to image textures and depth/disparity variations. WSSN firstly extracts both image texture features and disparity features by convolutional neural network and then utilizes the fully connected layers to conduct optimal window size selection. Experiments on an aerial image dataset show that our proposed method is capable of selecting appropriate matching windows for each pixel. It achieved the highest matching accuracy when compared with the matching results of a series of fixed matching windows and a state-of-the-art textured based window selection method. [ABSTRACT FROM AUTHOR]

Making an accurate diagnosis of nasopharyngeal carcinoma (NPC) disease is a challenging task that involves many parties such as radiology specialists often times need to delineate NPC boundaries on various tumor-bearing endoscopic images. It is a tedious and time-consuming operation exceedingly based on doctors and experience of radiologist. NPC has complex and irregular structures which makes it difficult to diagnose even by an expert physician. However, the diagnosis accuracy results of such methods are still insignificant and need improvement in order to manifest robust solution. The study aim is to develop and propose a new automatic classification of NPC tumor using machine learning techniques and feature-based decision-level fusion scheme from endoscopic images. We have implemented the fusion of the three image texture-based schemes (local binary patterns, the first-order statistics histogram properties, and histogram of gray scale) at the decision level and tested the performance of this scheme using the same experimental setup in the previous section for simple score-level fusion, but for comparison, We used the classifiers methods which are support vector machines (SVM), k-nearest neighbors' algorithm, and artificial neural network (ANN). The results demonstrate that the majority rule for decision-based fusion is outperformed considerably by the single best performing feature scheme (FFGF) for the SVM classifier, but for the ANN and KNN classifier it is significantly outperformed by each of the components features. The classifiers approaches were listed a high accuracy of 94.07%, the sensitivity of 92.05%, and specificity of 93.07%. [ABSTRACT FROM AUTHOR]

The objective of this study was to design and produce highly comfortable shoe products guided by a plantar pressure imaging data-set. In this research a plantar pressure optical imaging data-set based classification technology has been developed. In this paper, an improved local binary pattern (LBP) algorithm is used to retrieve texture-based features and recognize patterns from the data-set. A calculation model of plantar pressure imaging feature area is established subsequently. The data-set is classified by a neural network to guide the generation of various shoe-last surfaces. Firstly, the local binary mode is improved to adapt to the pressure imaging data-set, and the texture-based feature calculation is fully used to accurately generate the feature point set; hereafter, the plantar pressure imaging feature point set is then used to guide the design of last free surface forming. In the presented experiments of plantar imaging, multi-dimensional texture-based features and improved LBP features have been found by a convolution neural network (CNN), and compared with a 21-input-3-output two-layer perceptual neural network. Three feet types are investigated in the experiment, being flatfoot (F) referring to the lack of a normal arch, or arch collapse, Talipes Equinovarus (TE), being the front part of the foot is adduction, calcaneus varus, plantar flexion, or Achilles tendon contracture and Normal (N). This research has achieved an 82% accuracy rate with 10 hidden-layers CNN of rotation invariance LBP (RI-LBP) algorithm using 21 texture-based features by comparing other deep learning methods presented in the literature. [ABSTRACT FROM AUTHOR]

The research methodology consists of several stages to develop a noninvasive method of identifying the turgor of potato tubers during the storage. During the first stage, a graphic database (set of training data) has been created for selected varieties of potatoes. As a next step, special proprietary software called 'PID system' was used together with a commercial MATLAB package to extract parameters defining the digital image descriptors. This included: hue space models, shape coefficient and image texture. Thirdly, Artificial Neural Network (ANN) training was conducted with the use of Statistica and MATLAB tools. As a result of the analysis, a neural model has been obtained, which had the greatest classification features. [ABSTRACT FROM AUTHOR]

Splicing Image Forgery Identification controls and picture tampering with no proof being left behind has turned out to be exceptionally moderate and exceedingly utilized, because of presence of to a great degree intense altering apparatuses, for example, Adobe Photoshop. Along these lines, there has been a quick expansion of the digitally adjusted pictures on the Internet and prevailing press. The genuineness of a digital image experiences extreme dangers because of the ascent of capable digital image altering devices that effectively adjust the image substance without leaving any obvious hints of such changes. The splicing forgery should be possible by replicated a one/more area from source image and pasted into an objective picture to create a composite image called spliced image. In this manner, this sort of forgery is viewed as challenge issue difficult from tamper identification perspective. To influence the issue most exceedingly awful some to post preparing impacts, for example, blurring, JPEG compression, rotation and scaling perhaps presented in the spliced image. This study aims to perform quantification and data analysis following feature extraction using computational techniques to detect interesting textural and anatomical changes, these extracted features then can be used as a key to distinguish between different classes (Splicing Image and non-Splicing image). This paper presents a new framework to identify the spliced image by exploiting the image texture features, and to automatically identification of spliced images. To accurately identify the spliced image, the proposed solution uses different texture features to capture deferent texture related to the edge of the object and the colour features. We have combine these features to produce a good vector to describe the splice object. Two models have been proposed in this paper first combine the vectors of the three features and feed them to the ANN classifier. Second, use the majority voting of the result of three features to take the decision. This is followed by a ANN classifier; in this model we have trained the system with 30 training 20% validation 50% testing. We evaluated the effectiveness of the classification framework for identifying spliced images by compare the result with the manual label which is done by the people who have created the data sets. In this approach we have combine different features to capture different information and feed them to the neural network to identify the splicing image. The findings outcome from this study have shown an improved approach that automatically splicing image forgery identification. We have evaluated the splicing image forgery identification using the texture features. The identification accuracy in the technique used is about 98.06%., with 99.03% sensitivity and 96.07% specificity. [ABSTRACT FROM AUTHOR]

A novel supervised particle swarm optimization (S-PSO) classification algorithm is proposed for fault diagnosis. In order to improve the accuracy of fault diagnosis and obtain the global optimal solutions with a higher probability, two strategies, i.e. a hybrid particle position updating strategy and a fixed iteration interval intervention updating strategy, are designed to balance the effect of the local and the global search. These methods increase the diversity of particles, expand the particles ability of searching the entire solution space, and guide the particles adaptively jumping out of the local optimal area. Meanwhile, based on the shorter intra-class distance, longer inter-class distance and maximum classification accuracy of training samples, a fitness function is designed to constraint the output optimal class centers. Experimental results demonstrate that the proposed S-PSO classification algorithm can overcome the problems in the classical clustering algorithms, which only consider the similarity of data instead of their physical meanings. The comparison on GE90 engine borescope image texture feature classification is also conducted. The results show that the performance of S-PSO classification algorithm is robust. Its classification accuracy is higher than those of popular methods, including support vector machine (SVM), neural network, Bayesian classifier, and k-nearest neighbor (k-NN) algorithm. [ABSTRACT FROM AUTHOR]

This paper presents an integrated system for the automatic analysis of mammograms to assist radiologists in confirming their diagnosis in mammography screening. The proposed automated confirmatory system (ACS) can process a digitalized mammogram online, and generates a high quality filtered segmentation of an image for biological interpretation and a texture-feature based diagnosis. We use a serial of image pre-processing and segmentation techniques, including 2D median filtering, seeded region growing (SRG) algorithm, image contrast enhancement, to remove noise, delete radiopaque artifacts and eliminate the projection of the pectoral muscle from a digitalized mammogram. We also develop an entire-image texture-feature based classification method, by combining a Rough-set approach to extract five fundamental texture features from images, and then an Artificial Neural Network technique to classify a mammogram as: normal; indicating the presence of a benign lump; or representing a malignant tumor. Here, 222 random images from the Mammographic Image Analysis Society (MIAS) database are used for the offline ACS training. Once the system is tuned and trained, it is ready for the automated use for the analysis and diagnosis of new mammograms. To test the trained system, a separate set of 100 random images from the MIAS and another set of 100 random images from the independent BancoWeb database are selected. The proposed ACS is shown to be successful in confirming diagnosis of mammograms from the two independent databases. [ABSTRACT FROM AUTHOR]

We propose a novel multistage smoke detection algorithm based on inherent optical characteristics such as diffusion, color, and texture of smoke. Moving regions in a video frame are detected by an approximate median background subtraction method using the diffusion behavior of smoke. These moving regions are segmented by a fuzzy C-means (FCM) clustering algorithm that uses the hue and saturation components of moving pixels in the hue-saturation-intensity color space. A decision rule is used to select candidate smoke regions from smoke-colored FCM clusters. An object tracking approach is employed in the candidate smoke region to detect candidate smoke objects in the video frame, and image texture parameters are extracted from these objects using a gray level co-occurrence matrix (GLCM). The thirteen GLCM features are selected to constitute the feature vector by applying principal components analysis, resulting in high-accuracy smoke detection. Finally, a back propagation neural network is utilized as a classifier to discriminate smoke and nonsmoke using the selected feature vector. Experimental results using a standard experimental dataset of video clips demonstrate that the proposed approach outperforms state-of-the-art smoke detection approaches in terms of accuracy, making real-life implementation feasible. [ABSTRACT FROM AUTHOR]

In this paper, a new attempt has been made in the area of tool-based micromachining for automated, non-contact, and flexible prediction of quality responses such as average surface roughness ( R), tool wear ratio (TWR) and metal removal rate (MRR) of micro-turned miniaturized parts through a machine vision system (MVS) which is integrated with an adaptive neuro-fuzzy inference system (ANFIS). The images of machined surface grabbed by the MVS could be extracted using the algorithm developed in this work, to get the features of image texture [average gray level ( G)]. This work presents an area-based surface characterization technique which applies the basic light scattering principles used in other optimal optical measurement systems. These principles are applied in a novel fashion which is especially suitable for in-process prediction and control. The main objective of this study is to design an ANFIS for estimation of R, TWR, and MRR in micro-turning process. Cutting speed ( S), feed rate ( F), depth of cut ( D), G were taken as input parameters and R, TWR, MRR as the output parameters. The results obtained from the ANFIS model were compared with experimental values. It is found that the predicted values of the responses are in good agreement with the experimental values. [ABSTRACT FROM AUTHOR]

Many organisms rely on reedbed habitats for their existence, yet, over the past century there has been a drastic reduction in the area and quality of reedbeds in the UK due to intensified human activities. In order to develop management plans for conserving and expanding this threatened habitat, accurate up-to-date information is needed concerning its current distribution and status. This information is difficult to collect using field surveys because reedbeds exist as small patches that are sparsely distributed across landscapes. Hence, this study was undertaken to develop a methodology for accurately mapping reedbeds using very high resolution QuickBird satellite imagery. The objectives were to determine the optimum combination of textural and spectral measures for mapping reedbeds; to investigate the effect of the spatial resolution of the input data upon classification accuracy; to determine whether the maximum likelihood classifier (MLC) or artificial neural network (ANN) analysis produced the most accurate classification; and to investigate the potential of refining the reedbed classification using slope suitability filters produced from digital terrain data. The results indicate an increase in the accuracy of reedbed delineations when grey-level co-occurrence textural measures were combined with the spectral bands. The most effective combination of texture measures were entropy and angular second moment. Optimal reedbed and overall classification accuracies were achieved using a combination of pansharpened multispectral and texture images that had been spatially degraded from 0.6 to 4.8 m. Using the 4.8 m data set, the MLC produced higher classification accuracy for reedbeds than the ANN analysis. The application of slope suitability filters increased the classification accuracy of reedbeds from 71% to 79%. Hence, this study has demonstrated that it is possible to use high resolution multispectral satellite imagery to derive accurate maps of reedbeds through appropriate analysis of image texture, judicious selection of input bands, spatial resolution and classification algorithm and post-classification refinement using terrain data. [ABSTRACT FROM PUBLISHER]

This paper presents a system for automated, non-contact, and flexible prediction of surface roughness of end-milled parts through a machine vision system which is integrated with an artificial neural network (ANN). The images of milled surface grabbed by the machine vision system could be extracted using the algorithm developed in this work, in the spatial frequency domain using a two-dimensional Fourier transform to get the features of image texture (major peak frequency F, principal component magnitude squared value F, and the average gray level G). Since F1 is the distance between the major peak and the origin, it is a robust measure to overcome the effect of lighting of the environment. The periodically occurring features such as feed marks and tool marks present in the gray-level image can be easily observed from the principal component magnitude squared value F The experimental machining variables speed S, feedrate F, depth of cut D, and the response extracted image variables F, F, and G could be used as input data, and the response surface roughness R measured by Surfcorder SE-1100 (traditional stylus method) could be used as output data of an ANN ability to construct the relationships between input and output variables. The ANN was trained using the back-propagation algorithm developed in this work due to its superior strength in pattern recognition and reasonable speed. Using the trained ANN, the experimental result had shown that the surface roughness of milled parts predicted by machine vision system over a wide range of machining conditions could be got with a reasonable accuracy compared with those measured by traditional stylus method. Compared with the stylus method, the constructed machine vision system is a useful method for prediction of the surface roughness faster, with a lower price, and lower environment noise in manufacturing process. Experimental results have shown that the proposed machine vision system can be implemented for automated prediction of surface roughness with accuracy of 97.53%. The results are encouraging that machine vision system can be extended to many real-time industrial prediction applications. [ABSTRACT FROM AUTHOR]

In this paper, a modular system for medical image archiving, management, diagnosis support, and telematic cooperation is presented. The system provides digital imaging and communications in medicine (DICOM)-compatible tools for digital image processing and database management of medical images. The software features algorithms for preprocessing, manual or semi-automatic segmentation, automatic calculation of geometrical/size characteristics, and 3-D visualization of organs or selected regions of interest. Additionally, the system incorporates a database where patient data and information can be stored and retrieved. Access to the database is only permitted to authorized users. The user-friendly interface makes the software handy and accessible to clinicians, whereas the telematic components allow collaboration with remote experts. The pilot system incorporates a computer-aided diagnosis module aiming at providing support in the diagnosis of focal liver lesions from computed tomography images. [ABSTRACT FROM AUTHOR]

A multisensor data fusion algorithm based on a multilayer neural network is presented for sea ice classification in the winter period. The algorithm uses European Remote Sensing (ERS), RADARSAT synthetic aperture radar (SAR), and low-resolution television camera images and image texture features. Based on a set of in situ observations made at the Kara Sea, a neural network is trained, and its structure is optimized using a pruning method. The performance of the algorithm with different combinations of input features (sensors) is assessed and compared with the performance of a linear discriminant analysis (LDA)-based algorithm. We show that for both algorithms a substantial improvement can be gained by fusion of the three different types of data (91.2% for the neural network) as compared with single-source ERS (66.0%) and RADARSAT (70.7%) SAR image classification. Incorporation of texture increases classification accuracy. This positive effect of texture becomes weaker with increasing number of sensors (from 8.4 to 6.4 percent points for the use of two and three sensors, respectively). In view of the short training time and smaller number of adjustable parameters, this result suggests that semiparametric classification methods can be considered as a good alternative to the neural networks and traditional parametric statistical classifiers applied for the sea ice classification. [ABSTRACT FROM AUTHOR]

The goal of exemplar-based texture synthesis is to generate texture images that are visually similar to a given exemplar. Recently, promising results have been reported by methods relying on convolutional neural networks (ConvNets) pretrained on large-scale image datasets. However, these methods have difficulties in synthesizing image textures with non-local structures and extending to dynamic or sound textures. In this article, we present a conditional generative ConvNet (cgCNN) model which combines deep statistics and the probabilistic framework of generative ConvNet (gCNN) model. Given a texture exemplar, cgCNN defines a conditional distribution using deep statistics of a ConvNet, and synthesizes new textures by sampling from the conditional distribution. In contrast to previous deep texture models, the proposed cgCNN does not rely on pre-trained ConvNets but learns the weights of ConvNets for each input exemplar instead. As a result, cgCNN can synthesize high quality dynamic, sound and image textures in a unified manner. We also explore the theoretical connections between our model and other texture models. Further investigations show that the cgCNN model can be easily generalized to texture expansion and inpainting. Extensive experiments demonstrate that our model can achieve better or at least comparable results than the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

By extending the pulsed recurrent random neural network (RNN) discussed in Gelenbe (1989, 1990, 1991), we propose a recurrent random neural network model in which each neuron processes several distinctly characterized streams of "signals" or data. The idea that neurons may be able to distinguish between the pulses they receive and use them in a distinct manner is biologically plausible. In engineering applications, the need to process different streams of information simultaneously is commonplace (e.g., in image processing, sensor fusion, or parallel processing systems). In the model we propose, each distinct stream is a class of signals in the form of spikes. Signals may arrive to a neuron from either the outside world (exogenous signals) or other neurons (endogenous signals). As a function of the signals it has received, a neuron can fire and then send signals of some class to another neuron or to the outside world. We show that the multiple signal class random model with exponential interfiring times, Poisson external signal arrivals, and Markovian signal movements between neurons has product form; this implies that the distribution of its state (i.e., the probability that each neuron of the network is excited) can be computed simply from the solution of a system of 2 Cn simultaneous nonlinear equations where C is the number of signal classes and n is the number of neurons. Here we derive the stationary solution for the multiple class model and establish necessary and sufficient conditions for the existence of the stationary solution. The recurrent random neural network model with multiple classes has already been successfully applied to image texture generation (Atalay & Gelenbe, 1992), where multiple signal classes are used to model different colors in the image. [ABSTRACT FROM AUTHOR]

The need for a fast and robust method to characterize nanostructure thickness is growing due to the tremendous number of experiments and their associated applications. By automatically analyzing the microscopic image texture of MoS2 and WS2, it was possible to distinguish monolayer from few-layer nanostructures with high accuracy for both materials. Three methods of texture analysis (TA) were used: grey level histogram (GLH), grey levels co-occurrence matrix (GLCOM), and run-length matrix (RLM), which correspond to first, second, and higher-order statistical methods, respectively. The best discriminating features were automatically selected using the Fisher coefficient, for each method, and used as a base for classification. Two classifiers were used: artificial neural networks (ANN), and linear discriminant analysis (LDA). RLM with ANN was found to give high classification accuracy, which was 89% and 95% for MoS2 and WS2, respectively. The result of this work suggests that RLM, as a higher-order TA method, associated with an ANN classifier has a better ability to quantify and characterize the microscopic structure of nanolayers, and, therefore, categorize thickness to the proper class. [ABSTRACT FROM AUTHOR]

• Developed an efficient medical image super-resolution method based on deep learning to assist the examination of knee osteoarthritis. • Utiltzed the medical image super-resolution method of end-to-end learning to reduce the reconstruction time. • Combined with the advantages of shallow super-resolution reconstruction network, the cascading convolution kernel is used to deepen the network, which effectively improves the image quality. • Effectively solves the problem of low resolution and low definition of medical images. With the rapid development of medical imaging and intelligent diagnosis, artificial intelligence methods have become a research hotspot of radiography processing technology in recent years. The low definition of knee magnetic resonance image texture seriously affects the diagnosis of knee osteoarthritis. This paper presents a super-resolution reconstruction method to address this problem. In this paper, we propose an efficient medical image super-resolution (EMISR) method, in which we mainly adopted three hidden layers of super-resolution convolution neural network (SRCNN) and a sub-pixel convolution layer of efficient sub-pixel convolution neural network (ESPCN). The addition of the efficient sub-pixel convolutional layer in the hidden layer and the small network replacement consisting of concatenated convolutions to address low-resolution images but not high-resolution images are important. The EMISR method also uses cascaded small convolution kernels to improve reconstruction speed and deepen the convolution neural network to improve reconstruction quality. The proposed method is tested in the public dataset IDI, and the reconstruction quality of the algorithm is higher than that of the sparse coding-based network (SCN) method, the SRCNN method, and the ESPCN method (+ 2.306 dB, + 2.540 dB, + 1.089 dB improved); moreover, the reconstruction speed is faster than its counterparts (+ 4.272 s, + 1.967 s, and + 0.073 s improved). The experimental results show that our EMISR framework has improved performance and greatly reduces the number of parameters and training time. Furthermore, the reconstructed image presents more details, and the edges are more complete. Therefore, the EMISR technique provides a more powerful medical analysis in knee osteoarthritis examinations. [ABSTRACT FROM AUTHOR]

A low-light image enhancement method based on a deep symmetric encoder–decoder convolutional network (LLED-Net) is proposed in the paper. In surveillance and tactical reconnaissance, collecting visual information from a dynamic environment and accurately processing that data is critical to making the right decisions and ensuring mission success. However, due to the cost and technical limitations of camera sensors, it is difficult to capture clear images or videos in low-light conditions. In this paper, a special encoder–decoder convolution network is designed to utilize multi-scale feature maps and join jump connections to avoid gradient disappearance. In order to preserve the image texture as much as possible, by using structural similarity (SSIM) loss to train the model on the data sets with different brightness level, the model can adaptively enhance low-light images in low-light environments. The results show that the proposed algorithm provides significant improvements in quantitative comparison with RED-Net and several other representative image enhancement algorithms. [ABSTRACT FROM AUTHOR]

In the field of single image defogging, there are two main methods. One is the image restoration method based on the atmospheric scattering theory which can recover the image texture details well. The other is the image enhancement method based on Retinex theory which can improve the image contrast well. In practice, however, the former can easily lead to low contrast images; the latter is prone to losing texture details. Therefore, how to effectively combine the advantages of both to remove fog is a key issue in the field. In this paper, we have developed a physics based generative adversarial network (PBGAN) to exploit the advantages between those two methods in parallel. To our knowledge, it is the first learning defogging framework that incorporates these two methods and to enable them to work together and complement each other. Our method has two generative adversarial modules, the Contrast Enhancement (CE) module and the Texture Restoration (TR) module. To improve contrast in the CE module, we introduced a novel inversion-adversarial loss and a novel inversion-cycle consistency loss for training the generator. To improve the texture in the TR module, we introduced two convolutional neural networks to learn the atmospheric light coefficient and the transmission map, respectively. Extensive experiments on both synthetic and real-world datasets demonstrate that the proposed approach performs better than several state-of-the-art methods quantitatively and qualitatively. [ABSTRACT FROM AUTHOR]

Corals are crucial animals as they support a large part of marine life. The automatic classification of corals species based on underwater images is important as it can help experts to track and detect threatened and vulnerable coral species. However, this classification is complicated due to the nature of coral underwater images and the fact that current underwater coral datasets are unrealistic as they contain only texture images, while the images taken by autonomous underwater vehicles show the complete coral structure. The objective of this paper is two-fold. The first is to build a dataset that is representative of the problem of classifying underwater coral images, the StructureRSMAS dataset. The second is to build a classifier capable of resolving the real problem of classifying corals, based either on texture or structure images. We have achieved this by using a two-level classifier composed of three ResNet models. The first level recognizes whether the input image is a texture or a structure image. Then, the second level identifies the coral species. To do this, we have used a known texture dataset, RSMAS, and StructureRSMAS. • Creation of a new coral structure dataset. • Study the performance of promising CNNs in the classification of coral images. • Analyse data augmentation on the performance of the coral classification models. • Evaluate image enhancement techniques on the performance of the models. • Creation of a two-level model able to classify any coral image, texture or structure. [ABSTRACT FROM AUTHOR]

Purpose: Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms. Methods: A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a linear discriminant classifier to score the detected masses. For the DCNN-based CAD system, ROIs from five consecutive slices centered at each candidate were passed through the trained DCNN and a mass likelihood score was generated. The performances of the CAD systems were evaluated using free-response ROC curves and the performance difference was analyzed using a non-parametric method. Results: Before transfer learning, the DCNN trained only on mammograms with an AUC of 0.99 classified DBT masses with an AUC of 0.81 in the DBT training set. After transfer learning with DBT, the AUC improved to 0.90. For breast-based CAD detection in the test set, the sensitivity for the feature-based and the DCNN-based CAD systems was 83% and 91%, respectively, at 1 FP/DBT volume. The difference between the performances for the two systems was statistically significant (p-value < 0.05). Conclusions: The image patterns learned from the mammograms were transferred to the mass detection on DBT slices through the DCNN. This study demonstrated that large data sets collected from mammography are useful for developing new CAD systems for DBT, alleviating the problem and effort of collecting entirely new large data sets for the new modality. [ABSTRACT FROM AUTHOR]

The high false-positive recall rate is one of the major dilemmas that significantly reduce the efficacy of screening mammography, which harms a large fraction of women and increases healthcare cost. This study aims to investigate the feasibility of helping reduce false-positive recalls by developing a new computer-aided diagnosis (CAD) scheme based on the analysis of global mammographic texture and density features computed from four-view images. Our database includes full-field digital mammography (FFDM) images acquired from 1052 recalled women (669 positive for cancer and 383 benign). Each case has four images: two craniocaudal (CC) and two mediolateral oblique (MLO) views. Our CAD scheme first computed global texture features related to the mammographic density distribution on the segmented breast regions of four images. Second, the computed features were given to two artificial neural network (ANN) classifiers that were separately trained and tested in a ten-fold cross-validation scheme on CC and MLO view images, respectively. Finally, two ANN classification scores were combined using a new adaptive scoring fusion method that automatically determined the optimal weights to assign to both views. CAD performance was tested using the area under a receiver operating characteristic curve (AUC). The AUC = 0.793  ±  0.026 was obtained for this four-view CAD scheme, which was significantly higher at the 5% significance level than the AUCs achieved when using only CC (p = 0.025) or MLO (p = 0.0004) view images, respectively. This study demonstrates that a quantitative assessment of global mammographic image texture and density features could provide useful and/or supplementary information to classify between malignant and benign cases among the recalled cases, which may eventually help reduce the false-positive recall rate in screening mammography. [ABSTRACT FROM AUTHOR]

Abstract: The development of non-destructive methods for the evaluation of cereal grain varieties has important implications for the food processing industry. The described experiment investigated 11 varieties of spring and winter wheat of different quality class. The analysis was performed on images acquired from a flatbed scanner interfaced to a PC. Kernel images were digitalized at high resolution (2673 × 4031) with 24-bit depth and 400 dpi. The variables input into the statistical model were the textures of single kernel projections. Textures were computed separately for seven channels (R, G, B, Y, S, U, V). The results were examined with the application of discriminant analysis and neural networks. The accuracy of texture-based classification of 11 wheat varieties reached 100%. The experimental design which yielded the most satisfactory results comprised texture measurements from the combined area of 20 kernels and variables from seven channels input into the neural network. The final classification quality was not affected by the year of cultivation, moisture content or grain variety. [Copyright &y& Elsevier]