Search

This paper presents the prediction of aging state of oil impregnated insulating Kraft paper using textural properties. Under conditions of prolonged thermal stress, the insulating capacity of Kraft paper decreases due to carbonization and degradation of cellulose. To analyze these effects, a new accelerated aging kit is developed for the experiments. Four groups of samples are collected, and the morphology is examined. Texture feature is extracted for the samples using normalization and Gray-Level Co-occurrence Matrix (GLCM). Supervised (Support Vector Machine (SVM)) and unsupervised (K-means) machine intelligence methods are trained to classify images based on the feature. In addition, new samples are checked for feasibility of the prediction of aging state. The calculated training accuracy of SVM is 92.67%, and for k-means is 95.0%, on the other hand, the testing accuracy of SVM is 91.0%, and for k-means is 92.6%. Therefore, this machine intelligence-based methodology using image features is effective in classifying the aging state of the Kraft Insulating Paper. [ABSTRACT FROM AUTHOR]

The dirty degree of banknotes determines to some extent whether banknotes can continue to circulate. This paper proposes a whale optimization algorithm based multi-layer support vector machine (WOA-MLSVMs) dirty degree recognition method based on the texture characteristics of banknote images. Based on the contact image sensor to collect the double-sided reflection images of the banknotes under red, green, blue, infrared and ultraviolet light, as well as the transmission images under the green light and infrared light, 22 texture characteristic parameters of the banknotes image based on the gray-scale co-occurrence matrix (GLCM) are extracted to describe the visual characteristics of the banknotes dirty degree, such as energy, entropy and inertia, etc. The banknotes images are selected based on the dirty degree recognition results of MLSVMs to establish the full-spectrum banknote dirty degree recognition sample data set. Five essential dimension estimation methods and seventeen data dimension reduction methods are combined to determine the essential dimension and the optimal dimension reduction method. Finally, WOA-MLSVMs realizes the full-spectrum banknote dirty degree recognition and the simulation results show the effectiveness of the proposed strategy. [ABSTRACT FROM AUTHOR]

In recent years, with the advent of digital imaging technology, e.g., color printers and color scanners, it has become easier for counterfeiters to produce fake banknotes. The spread of counterfeit money causes loss to everyone involved in financial transactions. Therefore, an effective and reliable verification technique is necessary for successful and reliable financial transactions. This paper presents a cognitive computationbased technique for paper currency verification. In this regard, Scanning Electron Microscopy (SEM) and X-Ray Diffraction (XRD) analyses of counterfeit and genuine banknotes were performed. This experimentation confirmed that the materials used in preparation of genuine and counterfeit banknotes were totally different from each other. Based on these findings, a set of discriminative and robust features was proposed to re ect these differences in currency images. The proposed features represented characteristics of the materials of the banknote, such as printing ink, chemical composition, and surface coarseness. With these robust features, Support Vector Machines (SVMs) were employed for classification. In order to evaluate the performance of the proposed technique, experimentations were performed on a self-constructed dataset of Pakistani banknotes, comprised of 195 currency images, including 35 counterfeit banknotes. The results showed that the proposed system achieved 100% verification ability for properly captured images. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we describe the development of a system for evaluating an specific quality characteristic of recycled paper sheets using techniques of image analysis and pattern recognition. We call Bumpiness the phenomenon of interest, which is new in the literature on paper quality. This phenomenon is characterized by the appearance of macroscopic undulations on the paper sheet surface that may emerge shortly or some time after its production. We explore the detection and measurement of this defect by means of computer vision and statistical pattern recognition techniques that may allow early detection at the production site. Our goal is to give an scalar continuous measure of Bumpiness. We propose features computed from Gabor filter banks (GFB) and discrete wavelet transforms (DWT) for the characterization of paper sheet surface Bumpiness in recycled paper images. The starting point is to state the problem as a classification of the paper sheet images into two classes: low and high Bumpiness. In this setting we obtain, with both proposed texture modelling approaches (GFB and DWT), classification accuracies comparable to the agreement between human observers. The best performance is obtained using DWT features. Finally, we propose as the scalar index of Bumpines the fisher discriminant analysis (FDA) function defined on the space of the best features for the classification task. We perform an innovative validation process of this Bumpiness index, based on the ordering of random pairs of images, obtaining a very high agreement with the human observers. [Copyright &y& Elsevier]

Background: The cloud-penetrating and fog-penetrating capability of Synthetic Aperture Radar (SAR) give it the potential for application in forest fire progress monitoring; however, the low extraction accuracy and significant salt-and-pepper noise in SAR remote sensing mapping of the burned area are problems. Aims: This paper provides a method for accurately extracting the burned area based on fully exploiting the changes in multiple different dimensional feature parameters of dual-polarised SAR images before and after a fire. Methods: This paper describes forest fire progress monitoring using dual-polarisation SAR images combined with multi-scale segmentation and unsupervised classification. We first constructed polarisation feature and texture feature datasets using multi-scene Sentinel-1 images. A multi-scale segmentation algorithm was then used to generate objects to suppress the salt-and-pepper noise, followed by an unsupervised classification method to extract the burned area. Key results: The accuracy of burned area extraction in this paper is 91.67%, an improvement of 33.70% compared to the pixel-based classification results. Conclusions: Compared with the pixel-based method, our method effectively suppresses the salt-and-pepper noise and improves the SAR burned area extraction accuracy. Implications: The fire monitoring method using SAR images provides a reference for extracting the burned area under continuous cloud or smoke cover. This paper describes a method to monitor forest fire progress using dual-polarisation Synthetic Aperture Radar (SAR) images combined with multi-scale segmentation and unsupervised classification. We aimed to take full advantage of the many different dimensions of feature parameter changes caused by forest fires, relying on time-series dual-polarised SAR imagery to achieve burned area extraction and forest fire progress monitoring. [ABSTRACT FROM AUTHOR]

Background: Cancer pathology shows disease development and associated molecular features. It provides extensive phenotypic information that is cancer-predictive and has potential implications for planning treatment. Based on the exceptional performance of computational approaches in the field of digital pathogenic, the use of rich phenotypic information in digital pathology images has enabled us to identify low-level gliomas (LGG) from high-grade gliomas (HGG). Because the differences between the textures are so slight, utilizing just one feature or a small number of features produces poor categorization results. Methods: In this work, multiple feature extraction methods that can extract distinct features from the texture of histopathology image data are used to compare the classification outcomes. The successful feature extraction algorithms GLCM, LBP, multi-LBGLCM, GLRLM, color moment features, and RSHD have been chosen in this paper. LBP and GLCM algorithms are combined to create LBGLCM. The LBGLCM feature extraction approach is extended in this study to multiple scales using an image pyramid, which is defined by sampling the image both in space and scale. The preprocessing stage is first used to enhance the contrast of the images and remove noise and illumination effects. The feature extraction stage is then carried out to extract several important features (texture and color) from histopathology images. Third, the feature fusion and reduction step is put into practice to decrease the number of features that are processed, reducing the computation time of the suggested system. The classification stage is created at the end to categorize various brain cancer grades. We performed our analysis on the 821 whole-slide pathology images from glioma patients in the Cancer Genome Atlas (TCGA) dataset. Two types of brain cancer are included in the dataset: GBM and LGG (grades II and III). 506 GBM images and 315 LGG images are included in our analysis, guaranteeing representation of various tumor grades and histopathological features. Results: The fusion of textural and color characteristics was validated in the glioma patients using the 10-fold cross-validation technique with an accuracy equals to 95.8%, sensitivity equals to 96.4%, DSC equals to 96.7%, and specificity equals to 97.1%. The combination of the color and texture characteristics produced significantly better accuracy, which supported their synergistic significance in the predictive model. The result indicates that the textural characteristics can be an objective, accurate, and comprehensive glioma prediction when paired with conventional imagery. Conclusion: The results outperform current approaches for identifying LGG from HGG and provide competitive performance in classifying four categories of glioma in the literature. The proposed model can help stratify patients in clinical studies, choose patients for targeted therapy, and customize specific treatment schedules. [ABSTRACT FROM AUTHOR]

In order to accurately extract the characteristics of debris flow caused by group rainstorms, effectively identify the on-site information of debris flow, and provide a scientific basis for debris flow monitoring, early warning and disaster control, this paper proposes a method for extracting the characteristics of heavy rainstorm debris flow using multiregional ecological environment remote sensing. In the ecological environment where debris flows occur frequently, remote sensing data of heavy rainstorm debris flows are preprocessed using remote sensing technology, providing an important basis for the feature extraction of debris flows. The kernel principal component analysis method and Gabor filters are innovatively used to extract the spectral and texture features of rainstorm and debris flow remote sensing images, and the convolutional neural network structure is improved based on the open source deep learning framework, integrating multilevel features to generate debris flow feature maps. The improved convolution neural network is then used to extract the secondary features of the fusion feature map, and the feature extraction of heavy rainstorm debris flow is realized. The experiment shows that this method can accurately extract the characteristics of heavy rainstorm debris flow. Fused remote sensing images of debris flow effectively ameliorate the problem of insufficient informational content in a single image and improve image clarity. When the Gabor kernel function has eight different directions, the feature extraction effect of the debris flow image in each direction of the heavy rainstorm is the best. [ABSTRACT FROM AUTHOR]

Image segmentation is a significant step in image analysis and computer vision. Many entropy-based approaches have been presented on this topic. Among them, Tsallis entropy isone of the best-performing methods. In this paper, the surface defect images of galvanised steel sheets were studied. Atwo-dimensional asymmetric Tsallis cross-entropy image segmentation algorithm based on chaotic bee colony algorithm optimisation was used to investigate the segmentation of surface defects under complex texture backgrounds. On the basis of Tsallis entropy threshold segmentation, a more concise expression form was used to define the asymmetric Tsallis cross-entropy in order to reduce the calculation complexity of the algorithm. The chaotic algorithm was combined with the artificial bee colony (ABC) algorithm to construct the chaotic bee colony algorithm, so that the optimal threshold of Tsallis entropy could be quickly identified. The experimental results showed that compared with the maximum Shannon entropy algorithm, the calculation time of this algorithm decreased by about 58% and the threshold value increased by about (26%, 54%). Compared with the two-dimensional Tsallis cross-entropy algorithm, the calculation time of this algorithm decreased by about 35% and about 20% was improved in the g-axis direction only. Compared with the two-dimensional asymmetric Tsallis cross-entropy algorithm, the calculation time of this algorithm decreased by about 30% and the threshold values of the two algorithms were almost the same. The algorithm proposed in this paper can rapidly and effectively segment defect targets, making it a more suitable method for detecting surface defects in factories with a rapid production pace. [ABSTRACT FROM AUTHOR]

Palmprint has gained popularity as a biometric modality and has recently attracted significant research interest. The competition-based method is the prevailing approach for hand-crafted palmprint recognition, thanks to its powerful discriminative ability to identify distinctive features. However, the competition mechanism possesses vast untapped advantages that have yet to be fully explored. In this paper, we reformulate the traditional competition mechanism and propose a $\boldsymbol {C}$ omprehensive $\boldsymbol {C}$ ompetition Network (CCNet). The traditional competition mechanism focuses solely on selecting the winner of different channels without considering the spatial information of the features. Our approach considers the spatial competition relationships between features while utilizing channel competition features to extract a more comprehensive set of competitive features. Moreover, existing methods for palmprint recognition typically focus on first-order texture features without utilizing the higher-order texture feature information. Our approach integrates the competition process with multi-order texture features to overcome this limitation. CCNet incorporates spatial and channel competition mechanisms into multi-order texture features to enhance recognition accuracy, enabling it to capture and utilize palmprint information in an end-to-end manner efficiently. Extensive experimental results have shown that CCNet can achieve remarkable performance on four public datasets, showing that CCNet is a promising approach for palmprint recognition that can achieve state-of-the-art performance. Related codes will be released at https://github.com/Zi-YuanYang/CCNet. [ABSTRACT FROM AUTHOR]

Seabed image segmentation is an important product for a variety of fields including habitat mapping, geological surveys, mine countermeasures, and naval route planning. Developing a clustering algorithm that can both accurately segment and effectively generalize high-resolution imagery for different seabed types over large areas is challenging. In this paper, we evaluate the performance of a new unsupervised image segmentation algorithm. The method utilizes imagery derived features (intensity and texture) to identify clusters (different seabed types) in feature space while also encouraging local homogeneity. We demonstrate how spatially coherent k-means clustering can efficiently and accurately segment synthetic aperture sonar (SAS) images. Our experiments show that spatially coherent clustering can significantly increase segmentation accuracy relative to OpenCV k-means and ArcGIS Pro iterative self-organizing (ISO) clustering (up to 15% and 20%, respectively). [ABSTRACT FROM AUTHOR]

Aiming at the problem of multi-type tree division structure in coding unit (CU) division in versatile video coding (VVC), this paper proposed a fast division algorithm based on the orientation characteristics and spatial complexity of CU subblocks. Firstly, this paper used the overall texture complexity of CU to classify the current CU and filtered out the CU that is not divided. Then, this method used the difference in the characteristics of different division directions of subblocks to decide the CU division direction in advance. Finally, the algorithm used the complexity difference feature between the middle region and the edge region of CU to determine whether to skip the ternary tree (TT) division, which further reduces the number of candidate list division patterns. The experimental results show that compared with the official test platform VTM10.0, the encoder reduces the encoding time by 40.25% at the cost of increasing the average output bit rate by 1.12%, indicating that the algorithm can achieve shorter encoding time with less quality loss in versatile video coding. [ABSTRACT FROM AUTHOR]

Early detection and diagnosis of brain tumors are essential for early intervention and eventually successful treatment plans leading to either a full recovery or an increase in the patient lifespan. However, diagnosis of brain tumors is not an easy task since it requires highly skilled professionals, making this procedure both costly and time-consuming. The diagnosis process relying on MR images gets even harder in the presence of similar objects in terms of their density, size, and shape. No matter how skilled professionals are, their task is still prone to human error. The main aim of this work is to propose a system that can automatically classify and diagnose glioma brain tumors into one of the four tumor types: (1) necrosis, (2) edema, (3) enhancing, and (4) non-enhancing. In this paper, we propose a combined texture discrete wavelet transform (DWT) and statistical features based on the first- and second-order features for the accurate classification and diagnosis of multiclass glioma tumors. Four well-known classifiers, namely, support vector machines (SVM), random forest (RF), multilayer perceptron (MLP), and naïve Bayes (NB), are used for classification. The BraTS 2018 dataset is used for the experiments, and with the combined DWT and statistical features, the RF classifier achieved the highest average accuracy whether for separated modalities or combined modalities. The highest average accuracy of 89.59% and 90.28% for HGG and LGG, respectively, was reported in this paper. It has also been observed that the proposed method outperforms similar existing methods reported in the extant literature. [ABSTRACT FROM AUTHOR]

COVID-19 has been a devastating pandemic, causing serious and sometimes irreparable damages to body organs. The sooner the existence of this virus in the body is recognized, the more effective the treatments are. This early detection can break the transmission chain faster, reducing the burden of this disease on the society. Since there exist issues regarding the reliability of RT-PCR tests to diagnose COVID-19, examining chest radiographs, especially chest X-ray images, are recommended as well. In this paper, we propose a machine learning algorithm to automatically classify patients in the target groups of COVID-19, pneumonia, and normal, based on chest X-ray images. Our algorithm generates two complementary images from each raw image in the dataset, and only works on a 4-feature vector extracted from the gray level co-occurrence matrix of each image for the classification, based on the k-nearest neighbors algorithm. It can work robustly in the presence of limited data. The speed, simplicity, and the independence from large computational resources are of other advantages of the proposed algorithm. Despite the simplicity and speed, as the results show, the algorithm can compete tightly with the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Context: Due to the change and advancement in technology, day by day the internet service usages are also increasing. Smartphones have become the necessity for every person these days. It is used to perform all basic daily activities such as calling, SMS, banking, gaming, entertainment, education, etc. Therefore, malware authors are developing new variants of malwares or malicious applications especially for monetary benefits. Objective: Objective of this research paper is to develop a technique that can be used to detect malwares or malicious applications on the android devices that will work for all types of packed or encrypted malicious applications, which usually evade decompiling tools. Method: In the proposed approach, visualization method is used for the detection of malware. In the first phase, application files are converted into images and then in second phase, texture feature of images are extracted using Grey Level Co-occurrence Matrix (GLCM). In the last phase, machine learning classification algorithms are used to classify the malicious and benign applications. Results: The proposed approach is run on different datasets collected from various repositories. Different efficiency parameters are calculated and the proposed approach is compared with the existing approaches. Conclusion: We have proposed a static technique for efficient detection of malwares. The proposed technique performs better than the existing technique. [ABSTRACT FROM AUTHOR]

Within digital forensics, a notable emphasis is placed on the detection of the application of fundamental image-editing operators, including but not limited to median filters, average filters, contrast enhancement, resampling, and various other operations closely associated with these techniques. When conducting a historical analysis of an image that has potentially undergone various modifications in the past, it is a logical initial approach to search for alterations made by fundamental operators. This paper presents the development of a deep-learning-based system designed for the purpose of detecting fundamental manipulation operations. The research involved training a multilayer perceptron using a feature set of 36 dimensions derived from the gray-level co-occurrence matrix, gray-level run-length matrix, and normalized streak area. The system detected median filtering, mean filtering, the introduction of additive white Gaussian noise, and the application of JPEG compression in digital Images. Our system, which utilizes a multilayer perceptron trained with a 36-feature set, achieved an accuracy of 99.46% and outperformed state-of-the-art deep-learning-based solutions, which achieved an accuracy of 97.89%. [ABSTRACT FROM AUTHOR]

The detection of human motion is receiving more attention amongst researchers, and is important in several applications. However, the issue is to offer effective monitoring sensors amongst various platforms without diminishing privacy. The radar models are beneficial for detecting human motion due to their potential to detect targets from long ranges and work in all types of weather. This paper develops a technique for human motion classification using Impulse Radio Ultra Wide Band (IR-UWB) with a wall radar model. The goal is to devise a human motion classification framework using a Random Multimodal Deep Learning (RMDL), which is tuned by the proposed optimization algorithm. Here, the Ultra Wide Band (UWB) signals are employed in the gridding process to evaluate the grids. The grids are adapted for feature extraction wherein the Hilbert transform features and texture features, like Local Gradient Pattern (LGP) and Local Optimal Oriented Pattern (LOOP) are considered. These features are considered in RMDL for identifying human motion. The training of RMDL is done using the proposed Spotted Grey Wolf Optimizer (SGWO), which is obtained by combining Spotted Hyena Optimizer (SHO) and Grey Wolf optimizer (GWO). The developed SGWO-based RMDL offered effective performance with the highest accuracy of 0.956, smallest Mean square error (MSE) of 0.200, highest True negative rate (TNR) of 0.959, and highest true positive rate (TPR) of 0.956. [ABSTRACT FROM AUTHOR]

The identification of sedimentary structures in lithofacies is of great significance to the exploration and development of Paleogene shale in the Boxing Sag. However, due to the scale mismatch between the thickness of laminae and the vertical resolution of conventional wireline logs, the conventional lithofacies division method fails to realize the accurate classification of sedimentary structures and cannot meet the needs of reservoir research. Therefore, it is necessary to establish a lithofacies identification method with higher precision from advanced logs. In this paper, a method integrating the gray level co-occurrence matrix (GLCM) and random forest (RF) algorithms is proposed to classify shale lithofacies with different sedimentary structures based on formation micro-imager (FMI) imaging logging and elemental capture spectroscopy (ECS) logging. According to the characteristics of shale laminae on FMI images, GLCM, an image texture extraction tool, is utilized to obtain texture features reflecting sedimentary structures from FMI images. It is proven that GLCM can depict shale sedimentary structures efficiently and accurately, and four texture features (contrast, entropy, energy, and homogeneity) are sensitive to shale sedimentary structures. To accommodate the correlation between the four texture features, the random forest algorithm, which has been proven not to be affected by correlated input features, is selected for supervised lithofacies classification. To enhance the model's ability to differentiate between argillaceous limestone and calcareous mudstone, the carbonate content and clay content calculated from the ECS logs are involved in the input features. Moreover, grid search cross-validation (CV) is implemented to optimize the hyperparameters of the model. The optimized model achieves favorable performance on training data, validation data, and test data, with average accuracies of 0.84, 0.79, and 0.76, respectively. This study also discusses the application of the classification model in lithofacies and production prediction. [ABSTRACT FROM AUTHOR]

Cervical cancer is the common cancer among women, where early-stage diagnoses of cervical cancer lead to recovery from the deadly cervical cancer. Correct cervical cancer staging is predominant to decide the treatment. Hence, cervical cancer staging is an important problem in designing automatic detection and diagnosing applications of the medical field. Convolutional Neural Networks (CNNs) often plays a greater role in object identification and classification. The performance of CNN in medical image classification can already compete with radiologists. In this paper, we planned to build a deep Capsule Network (CapsNet) for medical image classification that can achieve high accuracy using cervical cancer Magnetic Resonance (MR) images. In this study, a customized deep CNN model is developed using CapsNet to automatically predict the cervical cancer from MR images. In CapsNet, each layer receives input from all preceding layers, which helps to classify the features. The hyper parameters are estimated and it controls the backpropagation gradient at the initial learning. To improve the CapsNet performance, residual blocks are included between dense layers. Training and testing are performed with around 12,771 T2-weighted MR images of the TCGA-CESC dataset publicly available for research work. The results show that the accuracy of Customized CNN using CapsNetis higher and behaves well in classifying the cervical cancer. Thus, it is evident that CNN models can be used in developing automatic image analysis tools in the medical field. [ABSTRACT FROM AUTHOR]

Marine oil spills are one of the major threats to marine ecological safety, and the rapid identification of oil films is of great significance to the emergency response. Marine radar can provide data for marine oil spill detection; however, to date, it has not been commonly reported. Traditional marine radar oil spill research is mostly based on grayscale segmentation, and its accuracy depends entirely on the selection of the threshold. With the development of algorithm technology, marine radar oil spill extraction has gradually come to focus on artificial intelligence, and the study of oil spills based on machine learning has begun to develop. Based on X-band marine radar images collected from the Dalian 716 incident, this study used image texture features, the BP neural network classifier, and threshold segmentation for oil spill extraction. Firstly, the original image was pre-processed, to eliminate co-channel interference noise. Secondly, texture features were extracted and analyzed by the gray-level co-occurrence matrix (GLCM) and principal component analysis (PCA); then, the BP neural work was used to obtain the effective wave region. Finally, threshold segmentation was performed, to extract the marine oil slicks. The constructed BP neural network could achieve 93.75% classification accuracy, with the oil film remaining intact and the segmentation range being small; the extraction results were almost free of false positive targets, and the actual area of the oil film was calculated to be 42,629.12 m2. The method proposed in this paper can provide a reference for real-time monitoring of oil spill incidents. [ABSTRACT FROM AUTHOR]

The H.266/Versatile Video Coding (VVC) standard poses a great challenge for encoder design due to its high computational complexity and long encoding time. In this paper, the fast partitioning decision of coding blocks is investigated to reduce the computational complexity and save the coding time of VVC intra-frame predictive coding. A fast partitioning algorithm of VVC intra-frame coding blocks based on gradient structure similarity and directional features is proposed. First, the average gradient structure similarity of four sub-coding blocks under the current coding block is calculated, and two thresholds are set to determine whether the current coding block terminates the partitioning early or performs quadtree partitioning. Then, for the coding blocks that do not satisfy the above thresholds, the standard deviation of the vertical and horizontal directions of the current coding block is calculated to determine the texture direction and skip unnecessary partitioning to reduce computational complexity. Based on the VTM10.0 platform, this paper evaluates the performance of the designed fast algorithm for partitioning within the VVC coding unit. Compared with VTM10.0, the encoding rate is improved by 1.38% on average, and the encoder execution time is reduced by 49.32%. The overall algorithm achieves a better optimization of the existing VVC intra-frame coding technique. [ABSTRACT FROM AUTHOR]

Medical Image Captioning (MIC), is a developing area of artificial intelligence that combines two main research areas, computer vision and natural language processing. In order to support clinical workflows and decision-making, MIC is used in a variety of applications pertaining to diagnosis, therapy, report production, and computer-aided diagnosis. The generation of long and coherent reports highlighting correct abnormalities is a challenging task. Therefore, in this direction, this paper presents an efficient F D T - D r 2 T framework for the generation of coherent radiology reports with efficient exploitation of medical content. The proposed framework leverages the fusion of texture features and deep features in the first stage by incorporating ISCM-LBP + PCA-HOG feature extraction algorithm and Convolutional Triple Attention-based Efficient XceptionNet ( C - T a X N e t ). Further, fused features from the FDT module are utilized by the Dense Radiology Report Generation Transformer ( D r 2 T ) model with modified multi-head attention generating dense radiology reports by highlighting specific crucial abnormalities. To evaluate the performance of the proposed F D T - D r 2 T extensive experiments are conducted on publicly available IU Chest X-ray dataset and the best performance of the work is observed as 0.531 BLEU@1, 0.398 BLEU@2, 0.322 BLEU@3, 0.251 BLEU@4, 0.384 CIDEr, 0.506 ROUGE-L, 0.277 METEOR. An ablation study is carried out to support the experiments. Overall, the results obtained demonstrate the efficiency and efficacy of the proposed framework. [ABSTRACT FROM AUTHOR]

Breast cancer is the most menacing cancer among all types of cancer in women around the globe. Early diagnosis is the only way to increase the treatment options which then decreases the death rate and increases the chance of survival in patients. However, it is a challenging task to differentiate abnormal breast tissues from normal tissues because of their structure and unclear boundaries. Therefore, early and accurate diagnosis and classification of breast lesions into malignant or benign lesions is an active domain of research. Over the decade, numerous artificial neural network (ANN)-based techniques were adopted in order to diagnose and classify breast cancer due to the unique characteristics of learning key features from complex data via a training process. However, these schemes have limitations like slow convergence and longer training time. To address the above mentioned issues, this paper employs a meta-heuristic algorithm for tuning the parameters of the neural network. The main novelty of this work is the computer-aided diagnosis scheme for detecting abnormalities in breast ultrasound images by integrating a wavelet neural network (WNN) and the grey wolf optimization (GWO) algorithm. Here, breast ultrasound (US) images are preprocessed with a sigmoid filter followed by interference-based despeckling and then by anisotropic diffusion. The automatic segmentation algorithm is adopted to extract the region of interest, and subsequently morphological and texture features are computed. Finally, the GWO-tuned WNN is exploited to accomplish the classification task. The classification performance of the proposed scheme is validated on 346 ultrasound images. Efficiency of the proposed methodology is evaluated by computing the confusion matrix and receiver operating characteristic (ROC) curve. Numerical analysis revealed that the proposed work can yield higher classification accuracy when compared to the prevailing methods and thereby proves its potential in effective breast tumor detection and classification. The proposed GWO-WNN method (98%) gives better accuracy than other methods like SOM-SVM (87.5), LOFA-SVM (93.62%), MBA-RF (96.85%), and BAS-BPNN (96.3%) [ABSTRACT FROM AUTHOR]

Automatic segmentation of land use and land cover from high resolution remote sensing imagery has been an essential research area in image processing for the past two decades. Timely and reliable information of land use and land cover is very much essential in administration for proper planning and decision making in various areas like agriculture, urban development, environment protection, etc. In this paper, a new algorithm ULCDSA (Unsupervised Linear Contact Distributions Segmentation Algorithm) is proposed for unsupervised segmentation of high resolution panchromatic data. Texture features extracted with the help of linear contact distributions and mathematical morphology are used in this paper. The proposed method has been implemented and tested on various panchromatic images of N.R.S.C's Cartosat-II data sets, Google earth images and on other aerial images. The results have then been compared with gray scale co-occurrence matrix algorithms and promising results have been obtained. [ABSTRACT FROM AUTHOR]

Detection and recognition of Diabetic Retinopathy (DR) at the early phase can prevent the risk of gradual damage in the retina and vision loss. Many works have been introduced for automatic DR recognition and diagnosis in recent years. To date, there are still some issues that are required to work on to improve the quality and the performance of automatic DR recognition systems. Therefore, this paper introduces a machine learning-based approach for DR diagnosis and recognition by proposing texture analysis features of the GLDM technique (Contrast, Angular Second Moment, Entropy, Mean, and Inverse Difference Moment) feature and feed-forward neural network classifier. The proposed method has achieved a recognition accuracy of 95% according to undertaken experiments and performance analysis. [ABSTRACT FROM AUTHOR]

To reduce the number of bruised mangoes at source, it is important to determine the different storage times of mangoes after mild bruise. In order to address this issue, a hyperspectral imaging combined with deep learning model was proposed. First, the average spectrum of the sample bruised area was extracted as spectral features, and then, the six eigenvalues of the most representative PC1 image were calculated as texture features based on the gray level co‐occurrence matrix. In order to find the optimal discriminative model, random forest (RF), partial least squares discriminant analysis (PLS‐DA), extreme gradient boosting (XGBoost), and convolutional neural network (CNN) models were built based on spectral features, texture features, and spectral features combined with texture features (Feature Fusion 1), respectively. The results showed that the best model discriminating model was based on CNN under Feature Fusion 1, with an overall accuracy of 90.22%. To reduce the redundant information and noise introduced by the full spectrum, uninformative variable elimination (UVE) and competitive adaptive reweighted sampling (CARS) algorithms were used to filter the spectral features. The screened spectral features were fused with texture features (Feature Fusion 2) and modeled again with RF, PLS‐DA, XGBoost, and CNN. The results showed that the optimal model for discriminating different storage times of mangoes after bruise was the CNN model based on feature fusion 2 (CARS), with an overall accuracy of 93.48%. In summary, this study shows that the spectral features combined with texture features can be used to effectively improve the model's discriminative results for different storage times of mango after mild bruise. Compared to other machine learning models, the CNN model in this paper achieves better results. It provides a theoretical basis for hyperspectral imaging combined with deep learning in discriminating different storage times of mangoes after mild bruise. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Technology (Natural Science) is the property of Chongqing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Current dynamic texture motion-based features are practically all pixel-based signatures, therefore making the recognition slow and favoring the quality over computational speed, which is unreasonable for nowadays time-sensitive applications. Consequently, the goal of this work is to ensure a balanced accuracy and computational speed of recognition by exploring block-based motion features that are more likely adaptable for fast characterization rather than pixel-based. In this paper, we proposed four fast and innovative block-based motion approaches that characterize dynamic texture raw videos without any further segmentation for recognition purposes. Their originality is to adopt block motion estimation fast algorithms and introduce the novel Axial Spatio-temporal Motion Vector Components to be analyzed statistically using customized space-time texture features (first-order, second-order, and higher-order). Experimentations and results yielded a satisfying equilibrium between recognition and computational speed on multiple Dynamic texture datasets: DynTex, DynTex++ and UCLA compared to the pixel-based techniques with a high reduction of calculation time. [ABSTRACT FROM AUTHOR]

Microsatellites are short strings of repeated sequences scattered throughout the human genome. Microsatellite Instability (MSI) is a phenomenon in which the length of microsatellites changes due to the insertion or deletion of repeated units in tumor tissues. MSI type gastric cancer often has unique molecular phenotypes and clinicopathological characteris- tics, and the instability of microsatellites determines whether gastric cancer patients respond well to immunotherapy. Therefore, preoperative detection of MSI status is of great significance for the formulation of treatment plans for gastric cancer patients. Traditional MSI detection methods require immunohistochemistry and genetic analysis, which not only require additional costs, but also are difficult to be extended to every patient in clinical practice. In this paper, image feature extraction technology and machine learning algorithm are applied to quantitative analysis of high-resolution histopathological images of gastric cancer patients to predict the MSI status of gastric cancer patients. The original data of 279 cases are obtained from the TCGA database. After pre-processing and up-sampling, 442 samples are obtained, and 445 quantitative image features are extracted from the histopathological images of each sample, including the first-order statistics, texture features and small wave characteristics of the images. Lasso regression is used to screen features and construct predictive labels (Risk-score) of gastric cancer MSI status, and the performance of predictive labels is verified through logistics classification model. Then, multivariate analysis is carried out in combination with the clinical characteristics of each patient, and personalized train diagram is constructed for MSI status prediction. The experimental results show that the prediction performance AUC value of the prediction label based on histone image texture features is 0.74, and the AUC value of the existing MSI prediction model based on histone image texture features is 0.73. Based on all samples, the AUC value of the MSI prediction model constructed by combining clinical features and Risk- score is 0.802, while the AUC value of the existing MSI prediction model combining clinical features and image features is only 0.752, compared with the existing methods, the MSI prediction model proposed in this paper has better prediction performance and can provide more valuable reference information for the clinical decision-making of gastric cancer patients. [ABSTRACT FROM AUTHOR]

Finger vein recognition is very important in the identity authentication, but its performance is affected significantly by noise. The widely used Conditional Generative Adversarial Nets (CGAN) de-noising algorithm without accurate texture constraints is easy to damage the texture features of the image. In this paper, we propose a finger vein de-noising algorithm based on Custom Sample-Texture Conditional Generative Adversarial Nets (CS-TCGAN). The proposed algorithm effectively protects the texture features while removing noise. Firstly, the proposed algorithm uses texture loss, adversarial loss, and content loss as constraints, which lead to a better de-noising performance on finger vein image with blurred texture.Secondly, in order to avoid the checkerboard artifacts effect caused by up-sampling in de-convolution process which results in the loss of the vein information, the dimension preserving structure is adopted in the generator network to minimize this problem. Lastly, the noise distribution of finger vein images obtained in the practical application has been investigated to generate the training dataset for obtaining a de-noising model with better generalization. Specifically, the training dataset has been established by combining Poisson noise, salt/pepper noise, Gaussian noise, and speckle noise. The experimental results illustrate that the performance of the proposed algorithm is better than the traditional filtering de-noising approaches and the widely used CGAN de-noising algorithms. [ABSTRACT FROM AUTHOR]

Glass Passivation Parts (GPP) wafer texture defects are one of the most important factors affecting the accuracy of wafer defect detection. Template matching has local errors and low efficiency, and deep learning requires many training samples. In the early stage, defect training sample sets cannot be provided. This paper discusses the design of an effective GPP wafer grain region texture defect detection algorithm using a sub-region one-to-one mapping. A set of standard wafer datum is selected as the reference of grain region segmentation detection, and then the standard wafer images and test GPP wafer images are automatically calibrated and segmented, respectively. Then, a series of pre-processes were performed to equalize the sizes of the two grain-region images. Then the grain region was divided into an equal number of rectangular sub-regions of the same size according to the measurement precision requirement. The correlation degree of each test sub-region is judged by the designed three-channel RGB gray-scale similarity decision functions. Experiments show that the algorithm successfully achieved the necessary calibration and segmentation for the grain region. Compared with the template and histogram matching algorithms, the proposed method does not require a training set, the detection accuracy is significantly improved and the detection efficiency is up to 29.74 times better on average using the proposed algorithm. [ABSTRACT FROM AUTHOR]