Search

Multilabel remote sensing image classification (MLRSIC) has received increasing research interest. Taking the co-occurrence relationship of multiple labels as additional information helps to improve the overall performance. However, current methods only focus on using it to constrain the final feature which is output from a convolutional neural network (CNN). On the one hand, these methods need to exploit the potential of label correlation in feature representation fully. On the other hand, they increase the label noise sensitivity of the system, resulting in poor robustness. In this paper, a novel method called 'Semantic Interleaving Global chaNnel Attention' (SIGNA) is proposed for MLRSIC. First, the label co-occurrence graph is obtained according to the statistical information of the training set and fed into a graph neural network (GNN) to generate optimal semantic feature representations of each label. Next, the semantic features are interleaved with visual features which are extracted by CNNs to guide the overall features of the input image transform from the original feature space to the semantic feature space with embedded label relations. Then, global attention triggered by semantic interleaving is used to emphasize visual features in important channels. Finally, to make SIGNA easier to use and more optimized, multihead SIGNA-based feature adaptive weighting networks are proposed as plug-in blocks to plug into any layers of a CNN. For remote sensing images, better classification performance can be achieved by inserting the plug-in blocks into the shallow layers of CNNs. We conducted extensive experimental comparisons on three data sets: UCM, AID and DFC15. Experimental results demonstrate that the proposed SIGNA achieves superior classification performance compared to state-of-the-art (SOTA) methods. Notes that the codes of this paper will be open to the community for reproducibility research. [ABSTRACT FROM AUTHOR]

The IETE Journal of Research's first issue of Volume 70 covers a wide range of topics in communications, electromagnetics, opto-electronics, computers and computing, control engineering, electronic circuits, devices and components, instrumentation and measurement, medical electronics, and power electronics. The articles in this issue discuss various subjects, including improved network performance in wireless sensor networks, machine learning-based classification of cervical cancer, optimal energy reduction in 5G communication systems, and secure data transmission in the Internet of Things. The issue also includes papers on photonic sensors, detection of diabetic retinopathy, optimisation models for home health care scheduling and wind-integrated power systems, and efficient task scheduling in cloud computing environments. One research paper introduces a smartwatch-based machine learning model for monitoring epilepsy and detecting seizures, achieving a high accuracy of 99.40% in cross-validation. The document provides a summary of articles published in the journal, with authors who are experts in their fields and provide experimental findings and simulations to support their research. [Extracted from the article]

This paper proposes a novel approach, BTC-SAGAN-CHA-MRI, for the classification of brain tumors using a SAGAN optimized with a Color Harmony Algorithm. Brain cancer, with its high fatality rate worldwide, especially in the case of brain tumors, necessitates more accurate and efficient classification methods. While existing deep learning approaches for brain tumor classification have been suggested, they often lack precision and require substantial computational time.The proposed method begins by gathering input brain MR images from the BRATS dataset, followed by a pre-processing step using a Mean Curvature Flow-based approach to eliminate noise. The pre-processed images then undergo the Improved Non-Sub sampled Shearlet Transform (INSST) for extracting radiomic features. These features are fed into the SAGAN, which is optimized with a Color Harmony Algorithm to categorize the brain images into different tumor types, including Gliomas, Meningioma, and Pituitary tumors. This innovative approach shows promise in enhancing the precision and efficiency of brain tumor classification, holding potential for improved diagnostic outcomes in the field of medical imaging. The accuracy acquired for the brain tumor identification from the proposed method is 99.29%. The proposed BTC-SAGAN-CHA-MRI technique achieves 18.29%, 14.09% and 7.34% higher accuracy and 67.92%,54.04%, and 59.08% less Computation Time when analyzed to the existing models, like Brain tumor diagnosis utilizing deep learning convolutional neural network with transfer learning approach (BTC-KNN-SVM-MRI); M3BTCNet: multi model brain tumor categorization under metaheuristic deep neural network features optimization (BTC-CNN-DEMFOA-MRI), and efficient method depending upon hierarchical deep learning neural network classifier for brain tumour categorization (BTC-Hie DNN-MRI) respectively. This paper proposes a novel approach, BTC-SAGAN-CHA-MRI, for the classification of brain tumors using a Self-Attention based Generative Adversarial Network (SAGAN) optimized with a Color Harmony Algorithm. Brain cancer, with its high fatality rate worldwide, especially in the case of brain tumors, necessitates more accurate and efficient classification methods. While existing deep learning approaches for brain tumor classification have been suggested, they often lack precision and require substantial computational time. The proposed method begins by gathering input brain MR images from the BRATS dataset, followed by a pre-processing step using a Mean Curvature Flow-based approach to eliminate noise. The pre-processed images then undergo the Improved Non-Sub sampled Shearlet Transform (INSST) for extracting radiomic features. These features are fed into the SAGAN, which is optimized with a Color Harmony Algorithm to categorize the brain images into different tumor types, including Gliomas, Meningioma, and Pituitary tumors. This innovative approach shows promise in enhancing the precision and efficiency of brain tumor classification, holding potential for improved diagnostic outcomes in the field of medical imaging. [ABSTRACT FROM AUTHOR]

Nowadays, Mobile Ad hoc Network (MANET) gains more attraction due to its seamless and dynamic use. The existing methods in MANET are plagued by some weaknesses, like higher computation time, undermining their real-time responsiveness. Whereas, lower accuracy in intrusion detection compromises the reliability of these systems, potentially leading to missed threats posing a challenge for effective security measures in dynamic MANET environments. To mitigate these weaknesses, an Auto Metric Graph Neural Network optimized with Woodpecker Mating Algorithm is proposed in this paper for detecting Network Layer Attack in MANET (AGNN-WMA-DNLA-MANET). Initially, the data are gathered from CIC-IDS 2019 dataset. The gathered data undergoes pre-processing to eliminate redundancy and missing values replacement using the Local Least Squares method. Afterward, the preprocessing output is fed to AGNN for classification, which classifies the attack as active, passive, or normal. Then, Woodpecker Mating Algorithm (WMA) is proposed to optimize the AGNN weight parameters to ensure the accurate classification. The proposed technique is executed in NS2 tool. The metrics, like accuracy, false positive rate (FPR) and attack detection rate (ADR), computation time, and RoC is analyzed. The proposed AGNN-WMA-DNLA-MANET approach provides 14.68%, 7.142%, and 4.65% higher accuracy for active attack, 15.19%, 2.77%, and 9.85% higher accuracy for passive attack and 38.18%, 12.02%, and 7.59% better accuracy for normal compared with existing methods, such as Dependable intrusion detection scheme under deep convolutional neural network (DCNN-MANET), deep learning-based intrusion detection scheme for MANET (AE-DNN-MANET), and Benchmarking of machine learning for anomaly based intrusion detection scheme (ML-MANET), respectively. [ABSTRACT FROM AUTHOR]

Environmental applications play a significant role in the ongoing research area of Polarimetric Synthetic Aperture Radar (PolSAR) image classification. In this paper, a new model is proposed for classifying PolSAR images and applied to a part of the Greater Cairo area in the Nile basin, South of Delta, Egypt. First, the proposed model performs data pre-processing by extracting the coherency and covariance elements noted as [T] and [C] matrices, respectively. Second, temporal convolutional networks (TCN) deep learning is used to extract the features from coherency and covariance elements and then train the model. Third, the SoftMax classifier is used to classify the PolSAR image. Finally, the proposed model is tested with evaluation metrics. The obtained results show that the proposed model can achieve high classification performances. [ABSTRACT FROM AUTHOR]

The Time Space Diagram (TSD) can abstractly represent multiple data sources and the macroscopic state of road traffic. However, the TSDs may be incomplete due to missing data, which seriously affects traffic management. Therefore, this paper proposed a Multi-Directional Recurrent Graph Convolutional Network (MDRGCN) for reconstructing TSDs and estimating missing traffic speeds given sparse data. We designed multi-directional RNN layers for scanning the TSDs from horizontal and vertical directions, which can fully exploit the contextual dependencies of the traffic information. In addition, our model includes graph convolution layers for mining potential spatial correlations in the TSDs. The performance of the model reconstructed from TSDs is validated on the NGSIM dataset. We also provided a comparison with other advanced methods, and the experimental results show that our method can perform well at both low and high missing rates, significantly outperforming the baseline methods. [ABSTRACT FROM AUTHOR]

In the bridge structure, the main bearing element is the spans, which are directly subject to random and continuous loads, environmental influences, ageing, etc. These factors are easy to lead to material deterioration. Therefore, this study proposes a novel approach to investigate the mechanical properties of the material. The span of the bridge is the selected structure for investigation. This study focuses on the general evaluation of mechanical factors consistent with reality, so the following main factors are to be carried out. First, the viscoelastic model of material will be applied to set up and solve the governing differential equation of the beams with material characteristics involving the elastic modulus (E) and the viscous coefficient (C). The viscoelastic model is different from the traditional elastic model due to its non-linearity, reflecting the actual state of the structure. Second, the random vibration signal-based Loss Factor function (LF) calculation using the Power Spectral Density (PSD) to detect changes in structures. LF is a feature representing changes in material properties, including elasticity and viscosity, and is suitable for many types of bridge structures. Also, the paper uses Cubic Interpolation (CI) to generate a surface representing the LF distribution. This interpolation results in surfaces with respect to the LF values distributed by frequencies and spectral amplitudes. Finally, the LF distribution-based material investigation using Convolutional Neural Network (CNN) is proposed with high performance and accuracy. This study applies the proposed method to several bridges in Ho Chi Minh City, Vietnam. It demonstrates that LF is highly suitable for bridge health monitoring. [ABSTRACT FROM AUTHOR]

The intelligent segmentation of high-resolution remote sensing (HRS) image, also called as dense prediction task for HRS image, has been and will continue to be important research in the remote sensing community. In recent years, the growing wave of artificial intelligence (AI) technology has introduced innovative paradigms to this domain, yielding outstanding results and overcoming many challenges with conventional segmentation techniques. This paper provides a comprehensive review of these intelligent segmentation methodologies, including traditional pattern recognition, convolution neural network (CNN)-based, and Transformer-based techniques. However, the explosive but incomplete development of intelligent segmentation techniques also poses more challenges for earth observation experts, the most of which is the technical interpretability. Consequently, we consider these segmentation techniques in the aspect of explainable artificial intelligence (XAI). Data-centric XAI thinks the practical applications of the segmentation model while model-centric XAI will facilitate the understanding of decision-making processes and the adjustment of structural features. Moreover, this review identifies novel research questions and provides constructive insights and recommendations to HRS image segmentation tasks, which may shed new light on the intelligent segmentation methods within the remote sensing image understanding community. [ABSTRACT FROM AUTHOR]

Anemia is a severe health condition commonly prevalent among women of reproductive age and children below five years. Screening patients before the condition becomes critical and can save many lives. World Health Organization (WHO) has set the "Global nutrition target 2025-anemia," aiming to reduce 50% of anemia cases among women of reproductive age. This target can be achieved through a time-efficient, cost-effective, and easy-to-use tool. Traditional testing methods require specific chemicals, machines, and equipment that are not available everywhere. It also requires the presence of nurses, laboratory workers, and doctors. These methods are costly, time-consuming, and produce biohazard waste, thus polluting the environment. We developed an Artificial Intelligence (AI)-based bot that can be used for screening people for anemia. The bot service is based on two models: a segmentation model to segment the Region of Interest (ROI) and a classification model to classify anemic cases from normal ones. To train the model, we have collected data from 160 anemic and 140 non-anemic persons. In this paper, we have explained the architecture of the models, all the training parameters, and their deployment on cloud services using the REAN chatbot service. We manage to reach an Intersection Over Union (IOU) score of 0.922 for the segmentation model; validation recall of 0.95 and validation accuracy of 0.9699 for the classification model. This system is easy to use and does not depend on the availability of comprehensive laboratory infrastructure or trained personnel and thus can enable screening of anemia in low-resource settings. [ABSTRACT FROM AUTHOR]

Hypergraph neural networks (HGNNs), extending the techniques of graph neural networks, have been applied to various fields due to their ability to capture more complex high-order node relationships. However, for hyperspectral image (HSI) classification tasks, previous HGNN-based works usually constructed hypergraphs using pixels as nodes, resulting in massive computational costs. Meanwhile, pixel-level personalized features are required for HSI classification. To achieve high efficiency and accuracy simultaneously, this paper presents a fast hypergraph neural network with detail preservation (DPFHNet) for HSI classification. It constructs hypergraphs at the superpixel level to reduce time consumption and supplement pixel-level detail features through a classification refinement module. This framework contains multiple stages. Firstly, its main stage is designed with HGNNs from a superpixel viewpoint rather than pixels, providing a fast strategy to capture high-order complex relationships of multiple homogeneous irregular regions. After that, auxiliary stages based on convolutional neural networks are integrated into the main stage, which adopts a hierarchical design and attempts to acquire pixel-level spatial-spectral information before the hypergraph feature extraction of the main stage, assisting in learning more valuable features. Finally, a classification refinement module is constructed, which generates pixel-level detail features to refine the superpixel-level features obtained by HGNN. Experiments on three datasets illustrate that DPFHNet achieves competitive results and efficiency compared to advanced methodologies. [ABSTRACT FROM AUTHOR]

Advances in hardware capabilities and big data technologies over the past decade have enabled the application of deep learning techniques to address the challenge of super-resolution in remote-sensing images. Recently, while deep-learning-based methods have outperformed traditional methods, the abundance of information in remote-sensing images creates an imbalance between performance and computational resource consumption in current deep-learning-based methods. This paper introduces a stable super-resolution algorithm based on parameter reconstructive optimization to address these issues. First, based on the stable super-resolution generative adversarial network (SSRGAN), the algorithm employs a generator with networks using residual connections to reconstruct images with enhanced resolution. Next, it extracts content, adversarial, and regularization losses using the discriminator from a stable super-resolution generative adversarial network, which in turn guides the training of the network. Finally, upon completion of training, structural re-parameterization is conducted to optimize the multi-branch trained network into a plain inference network. This inference network can serve as an individual new generator. The results of qualitative and quantitative experimental comparisons with the models Bicubic, super-resolution convolutional neural network, very deep super-resolution convolutional network, deep recursive residual network, super-resolution generative adversarial network (SRGAN), enhanced SRGAN, and stable SRGAN on the Gaofen-5 AHSI satellite dataset suggest that this algorithm achieves improved evaluation indices with a 4× magnification ratio, reaching a peak signal-to-noise ratio of 30.7207 dB and structural similarity index measure of 0.8178. Compared with the trained but unconverted generator, which can also work independently, implementing re-parameterization results in approximately a 10% reduction in the number of parameters, indicating lower resource consumption, while the reconstruction effect is minimally influenced. Furthermore, the super-resolution results exhibit richer detail, increased contrast, and better scene adaptability. [ABSTRACT FROM AUTHOR]

Automatic building detection from remote sensing images holds significant applications in various domains, including cartography, land-use change detection, urban planning, and 3D intelligent city construction. Driven by the innovation of deep learning technology, convolutional neural networks (CNNs) have emerged as a powerful framework for automatic interpretation, exhibiting hierarchical feature generation and robust representation for building extraction. However, the inherent computational limitations of convolutional operators hinder the acquisition of an effective global receptive field. Recent advancements in deep learning have witnessed the adoption of the multi-head self-attention (MHSA) mechanism by Transformer, presenting the superiority of global context construction. Nonetheless, this network structure encounters difficulties in effectively modelling two-dimensional spatial structure features with scale variation in semantic segmentation and often has huge parameters with high algorithm complexity. To address the above problems, this paper proposed a hybrid deep neural network synergistically harnessing the superiority of CNN and Transformer to extract buildings from remote sensing images. Meanwhile, an improved MHSA mechanism is developed to facilitate the construction of multiscale global representations, concurrently reducing computational complexity. Furthermore, the cross-level MHSA modules the semantic correlation between the deep and shallow-level features to refine spatial fine-grained information. The proposed method was comprehensively evaluated on two building datasets via quantitative analysis and visualization. The experimental results confirmed that the developed network and modules outperform other state-of-the-art methods and can effectively improve the accuracy (93.4% F1-score on the WHU dataset, 89.38% F1-score on the CAMB dataset) with high efficiency for building extraction. [ABSTRACT FROM AUTHOR]

Many types of objects are crowdedly distributed on the surface of remote sensing scenes. Global semantics with object mixture and small training samples potentially cause difficulties in the classification of remote sensing scenes. The description of them often requires crucial parts of the scenes and corresponding discriminative features, especially as the convolutional neural network (CNN) goes deeper. In this paper, we propose a novel Location-Aware Multi-code Generator network (LAM-GAN) that incorporates multiple latent codes as the input to a generator network. This network is designed to train from scratch and recover most details of the input (real) image in a principled way. Meanwhile, multiple latent codes are reversely updated using K cluster centres located by the subsequently proposed part co-location module. By doing so, the global features of the real-fake image pair and part-level features are stacked and fed to a joint part classification network for discriminative classification. This approach makes it easier to induce the semantic concepts in a remote sensing scene. With this formulation, our approach generates an internal compact representation of the scene and enables weakly supervised part co-localization. The proposed method provides a unified framework for not only generating high-quality fake images but also facilitating the remote sensing scene classification task. We evaluated LAM-GAN on several benchmark datasets, and the experiment results demonstrate that the proposed method is more effective than previous state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The rapid deployment of 5G networks necessitates innovative solutions for efficient and dynamic resource allocation. Current strategies, although effective to some extent, lack real-time adaptability and scalability in complex, dynamically-changing environments. This paper introduces the Dynamic Resource Allocator using RL-CNN (DRARLCNN), a novel machine learning model addressing these shortcomings. By merging Convolutional Neural Networks (CNN) for feature extraction and Reinforcement Learning (RL) for decision-making, DRARLCNN optimizes resource allocation, minimizing latency and maximizing Quality of Service (QoS). Utilizing a state-of-the-art “5G Resource Allocation Dataset”, the research employs Python, TensorFlow, and OpenAI Gym to implement and test the model in a simulated 5 G environment. Results demonstrate the effectiveness of DRARLCNN, showcasing an impressive R2 score of 0.517, MSE of 0.035, and RMSE of 0.188, surpassing existing methods in allocation efficiency and latency. The DRARLCNN model not only outperforms existing methods in allocation efficiency and latency but also sets a new benchmark for future research in dynamic 5G resource allocation. Through its innovative approach and promising results, DRARLCNN opens avenues for further advancements in optimizing resource allocation within dynamic 5G networks. [ABSTRACT FROM AUTHOR]

Real-time monitoring of transient stability has gained significant importance in the power system. This paper proposes an advanced hybrid Transient Stability Assessment (TSA) monitoring system, composed of a Convolutional Neural Network (CNN) and Random Forest (RF) classifier which captures the PMU measurements in real-time. CNN utilizes heatmaps for detecting the transient stability status which is hereby used as a feature extractor for the RF classifier. The IEEE-14 and the IEEE-39 bus systems are used as case studies in MATLAB. A wind turbine with a doubly-fed induction generator is also installed to inspect its impact on the power system stability. The efficiency of the system is tested under different contingency scenarios by implementing various faults on distinct locations under dynamic loading conditions. The test results verified that the proposed hybrid CNN-RF (Convolutional Neural Network-Random Forest) model exhibited excellent results in terms of monitoring accuracy, higher precision, and performance in the presence of noise. [ABSTRACT FROM AUTHOR]

In recent years, numerous hyperspectral image classification methods based on convolutional neural network (CNN) have been proposed in recent years. However, the majority of CNN-based classification models fail to take into account the semantic gap between shallow and deep features, and instead, just sum up or splice these two types of features, which leads to subpar feature fusion. To address the above problems, the paper proposes a hyperspectral image classification method based on Narrowing Semantic Gap Convolutional Neural Network (NSGCNN). We developed the feature fusion sub-network, which consists of Feature Semantic Enhancement Module (FSEM), Spatial Detail Supplement Module (SDCM), and Cross-layer Feature Fusion Module (CFFM). FSEM extracts rich global semantic information based on multi-scale dilated convolution. Reverse attention mechanism is introduced in the SDCM to complement deep-layer features with fine spatial features like edges and textures. The fusing of hierarchical features is then achieved using CFFM. Additionally, we can regulate the model's parameter count by regulating the number of output feature maps. Experimental results on three benchmark hyperspectral datasets demonstrate that NSGCNN outperforms other state-of-the-art methods in terms of classification performance and significantly outperforms them in terms of model complexity. [ABSTRACT FROM AUTHOR]

PV(photovoltaic) systems have been deployed more in the recent years to support green energy generation. In the recent era, grid tied PV has been sparked by the emergence of the electricity market and offers an alternative solution to traditional fossil fuel-based electricity generation. A few of the potential impacts of solar PV on grid are false tripping of feeders, unwanted tripping, unnecessary islanding, and blind protection. In this research paper, different artificial intelligence techniques are tested in order to overcome the protection challenges. The PSCAD/EMTDC software package is used to analyze a section of the power system, and an algorithm has been constructed using Python v3.8 and MATLAB 2018. On an IEEE-9 BUS power system connected to a PV source, the Artificial Neural Network, Naive Bayes, Support Vector Machine, Random Forest, and Convolution Neural Network (CNN) algorithms are implemented to classify the faults. The suggested methods are proven effective for both in-zone and out-of-zone problems on power lines interconnected with solar park. The proposed techniques have been validated using total 16,320 internal and external fault cases with a wide range of system parameters alteration. In the proposed system, the effectiveness of several machine learning and deep learning techniques is compared. The obtained results demonstrate that CNN provides greater accuracy in the presence of a PV source, but at the same time, it is appropriate for a large number of data sets. The fault classification accuracy acquired is adequate and demonstrates the adaptability of the proposed approach. [ABSTRACT FROM AUTHOR]

Potato is one of the most nutritious foods rich in fiber and essential vitamins and minerals. However, potato plants may suffer from diseases such as early blight, late blight, septoria leaf spot or skin diseases such as greening, silver scurf, common scab, black rot etc. These diseases may lead to huge losses in the potato yield annually. Hence, an early prediction of such diseases is required so that they can be controlled at an early stage. Several machine learning, IoT based and image processing techniques have been utilized for this purpose, and this paper aims to survey all such methods and prediction models. A total of 116 studies were analyzed, and out of them, 77 relevant studies belonging to (60%) journals, (34%) conferences and (6%) other sources have been surveyed. It has been observed that most of the selected studies (64%) have used RGB image data (proprietary or PlantVillage dataset). Also, majority of the studies (48%) have worked with imbalanced data. Late blight and early blight are the two diseases that have gained the most attention by the researchers, being included in 58% and 34% of the studies. Moreover, feature extraction/selection has been used only with image data (RGB and hyperspectral). Convolutional Neural Network is the most widely used classifier, followed by Support Vector Machines. This review will help the researchers gain an insight into the work done for disease detection in potato crop and what techniques or diseases need intervention so that yield losses can further be minimized. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural network (CNN) and Transformer have been successfully applied in hyperspectral image (HSI) classification. Despite their success, some challenges persist. Popular HSI classification methods typically employ only a single patch as the network input, which may not provide robust information for various datasets or within different scenarios of a specific dataset. Large-scale patches are usually conducive to classifying large-scale scenes, while small-scale patches excel at classifying boundary locations. In addition, the majority of methods do not distinguish between the central pixel and its adjacent pixels, although the information from the central position is crucial. In this paper, a novel double-branch hierarchical partial convolution network (DBHP-Net) is proposed to address the aforementioned challenges. Concretely, DBHP-Net employs a new convolution, named TConv, to emphasize mining the spatial-channel features of the central position. Then, the TConvFormer Block is built by adopting the advanced engineering techniques of Transformer. Furthermore, we construct our model in a hierarchical style. The double-branch structure with large-scale and small-scale patches provides complementary information from the dataset. Experimental results based on randomly selected samples and spatially disjoint samples of two benchmark datasets demonstrate that our method outperforms other state-of-the-art (SOTA) methods. [ABSTRACT FROM AUTHOR]

This paper devises a technique for diagnosing and classifying osteoporosis using the femur bone's X-ray images. The devised approach uses the following phases: image acquisition, pre-processing, segmentation of femur boundary, measurement of femoral geometry, feature mining, and osteoporosis categorization. Here, femur boundary segmentation is done using the Deep Contour Aware Network (DCAN), which is trained by the Chronological Artificial Ecosystem Optimization (CAEO) algorithm. Then, the femoral geometry is measured and the most relevant features in the image are identified. Further, osteoporosis is classified using the Convolutional Neural Network (CNN) with Transfer Learning (TL), where the CNN is employed with the hyperparameters of the pre-trained model, LeNet. Here, the Chronological Artificial Ecosystem Tasmanian Devil Optimization (CAETDO) algorithm optimizes the classification process. The CAETDO-CNN-TL is examined based on metrics and have the values of 0.910, 0.927, 0.928, and 0.904 for precision, True Negative Rate (TNR), True Positive Rate (TPR), and accuracy, correspondingly. [ABSTRACT FROM AUTHOR]

Hyperspectral images (HSIs) have transformed the field of remote sensing by providing researchers with a wealth of information about the Earth's surface. However, analyzing these images can be an overwhelming task due to the presence of overlapping areas, nested regions, and large intra-class variability. Hyperspectral image classification (HSIC) is a crucial part of identifying the various land cover classes present in hyperspectral images. In order to enhance the accuracy of HSIC, researchers utilize the potential of three-dimensional convolutional neural networks (3D-CNNs). With the ability to influence both the spectral and spatial data present in HSIs, 3D-CNNs provide a promising solution to overcome the challenges associated with HSIC. In this paper, a new method for key band selection is proposed to improve the performance of 3D-CNN model. The proposed method selects the most relevant key bands based on Walsh-Hadamard kernel strength features. These key bands are then used to extract overlapping 3D spatial patches, which serve as input to the proposed 3D-CNN model. To evaluate the performance of the 3D-CNN model six standard benchmark datasets are used. The effectiveness of the proposed method improves the performance of 3D-CNN for HSIC. [ABSTRACT FROM AUTHOR]

The purpose of pan-sharpening is to generate high-resolution multispectral (HRMS) images by combining multispectral images (MS) and panchromatic images (PAN), which have become an important part of remote sensing image processing. Therefore, how to better extract complete feature information from MS images and PAN images has become the focus of our attention. In this paper, we propose a new pansharpening method, called Transformer-based Dual-path cross fusion network (TDF) for Pan-sharpening remote sensing images, which aims to extract the spatial details of PAN images while maintaining the spectral fidelity of MS images. The whole network structure can be divided into two parts: in the encoder part, we adopt the Swin-Transformer module for the downsampling operation, which expands the sensory field of the network to the feature map, and then extracts the global information. However, since the Swin-Transformer module is not good at extracting pixel-level details, we introduce the Base Feature Extraction (BFE) and the Invertible Neural Network Block (INNB) modules for the interaction between the local and the global feature information. we also introduce the Edge-Enhancement Block (EEB) to further enhance the feature extraction at multi-scales during the image fusion process. In the decoder section, we once again employ the Swin Transformer module for downsampling. After the convolution operation and activation function, we utilize the Sub-Pixel Convolutional Neural Network for upsampling to generate the ultimate high-resolution multispectral images. Simulation experiments and real experiments are conducted on QuickBird (QB) and WorldView2 (WV2) datasets, which demonstrated our method are superior to the current methods. [ABSTRACT FROM AUTHOR]

Because of the Internet of Things (IoT) and wearable devices, information obtained is subject to cyber attacks, making intrusion detection an essential component. The increased number of attacks on IoT devices exposes them to ongoing exploitation and data theft. Mirai, denial of Service (DoS), scan, etc are the major types of attacks conducted on IoT. Traditional attack detection algorithms have more disadvantages like lack of adaptability in all scenarios, low local minima computational complexity, etc. Thus developing a method that is suitable for all environments and has good convergence speed is taken as the motivation of this study. This paper presents a novel Symbiotic Organisms Search Algorithm (SOSA) optimized Faster region-based convolutional neural network (FRCNN) for attack detection and data type classification. The data to be stored in the cloud is analyzed whether they have restricted access, confidential, or unclassified. Depending upon the variety of data, the Optimal homomorphic encryption (OHE) approach is provided for security. The key selection process is done by an Elastic Collision Seeker Optimization Algorithm (ECSOA) which offers secure access to the data owner and prevents the data from unauthorized access. The model is optimized using the algorithm which improves the feature extraction performance. The effectiveness of the proposed model in attack detection is evaluated using two datasets namely the CSE-CIC-ID2018 dataset and the IoTID20 dataset. When compared to the existing techniques, the proposed model offers optimal performance in terms of multiclass attack detection. [ABSTRACT FROM AUTHOR]

Various prediction approaches regarding liver diseases have been developed. Still, they are expensive and more complex. This work aims to design an effective method for identifying liver diseases at earlier stages. This paper presents a modified Mask-regionalconvolutional neural network (MRCNN) architecture for non-invasive liver disease prediction. The Pelican Optimization Algorithm (POA) is used to balance the bounding box regression and mask branching training losses of the RCNN model. The liver disease features are extracted from three datasets namely Indian liver patient records, Hepatitis C, and Cirrhosis Prediction datasets. The POA-modified MRCNN model is mainly used to identify the interrelationship that exists between different laboratory measurements and diagnoses. The efficiency of the proposed model is compared with different state-of-art methods such as Opposition-based Laplacian Equilibrium Optimizer, Adaptive Hybridized Deep CNN, SVM, and Tree-based classifiers in terms of Accuracy, Precision, Recall, F-measure, and Mathews Correlation Coefficient (MCC). The proposed model offers an MCC value of 94.89, an accuracy of 98%, a precision of 96.2%, an F-measure of 97.3%, and a recall of 95% respectively. The results demonstrate the efficiency of the proposed model in predicting liver disease at an early stage via automatic screening and minimizing the burden of physicians. [ABSTRACT FROM AUTHOR]

This paper presents a new method for teeth and anomalies detection and classification using Faster Region Convolutional Neural Network with Deep Learning. Four classes of teeth and two classes of teeth anomalies are used for the classification by using Orthopantomogram radiography images as input. Using the Regional Proposal Network (RPN) and Object Detection Network (ODN), the detection of teeth objects has been made possible which replaces the manual segmentation of each individual tooth from the set of teeth signals thereby making the whole system more efficient. Overfitting is avoided by using the Dropout technique and thus improves the accuracy of the system. The model is trained and tested with the input samples and also compared with the ground truth and it achieves an accuracy of 92% for detection and 99.72% for classification. [ABSTRACT FROM AUTHOR]

Deep learning technologies like convolutional neural networks have recently become popular in the field of image denoising. A combined discrete wavelet transform (DWT) and denoising CNN (DnCNN)-based technique suggested in this paper can be used to adaptively boost image contrast. Images were captured in a wide variety of lighting circumstances, which means that not all of them are of the highest quality. As a result of the limited dynamic range of the pixel values, the overall image quality suffers when a shot is taken in low light. We employ a lossy method based on a structural similarity index to keep as much of the original texture in the images as feasible. Current advancements in image denoising, such as energy band analysis, very deep architecture, learning algorithms, dense-sparse-dense training, and regularization approaches, were applied in the hybrid DWT-Denoising convolutional neural networks. DnCNN eliminates the latent information in the hidden layers by design, resulting in a clean image. The DWT-DnCNN delivers SSIM values that are 2% higher than those of the standard RIDNet. Our hybrid DWT-DnCNN-based contrast enhancement strategy surpasses existing methods, according to simulation outcomes. [ABSTRACT FROM AUTHOR]

The rapid development in computer technology plays an essential role in the research works for performing fast and accurate identification of flower species through the processing of flower images with the support of mobile devices that seeks more attention in the research areas. It is highly significant for maintaining the sensitivity of ecological balance, and therefore, the image-processing approaches have provided improved outcomes in recent days. To rectify these existing problems and improve the accuracy attains the flower classification, the CNN variants of ensemble techniques can be used for the dynamic ensemble selection of the CNN networks. This paper develops an enhanced ensemble deep learning-based flower classification model to get efficient classification results with optimal models. The pre-processing is done through the Contrast Limited Adaptive Histogram Equalization (CLAHE) and filtering techniques. The pre-processed images are considered for the optimal pattern extraction phase, where optimal hybrid patterns are extracted from the Local Binary Pattern (LBP) and Local Vector Pattern (LVP). Here, the optimal hybrid pattern extraction phase contains the optimization in it using the enhanced heuristic algorithm named Improved Rat Swarm Optimizer (IRSO). The flower classification is performed with an Adaptive Dynamic Ensemble Transfer learning-based Convolutional Neural Network (ADET-CNN). Here, the optimal models are selected with the support of the same RSO algorithm. The experimental results show a better efficacy of the developed flower classification framework. [ABSTRACT FROM AUTHOR]

Predicting landslide hazards benefits geological disaster prevention and control. A novel cellular automaton (CA) integrating spatial case-based reasoning (SCBR), namely SCBR-CA, is proposed in this paper to predict landslide hazards at a local scale. The proposed model not only extracts spatial scene features for computations but also achieves dynamic prediction, which means that only one input is needed to obtain continuous predictions. Experiments were performed in Lushan, Sichuan, China. After using a convolutional neural network (CNN) to obtain the initial static landslide hazard zoning results, the landslide hazard zoning results for 2016–2025 were predicted with the SCBR-CA model. For comparison, a CA combined with a CNN (CNN-CA), was introduced. The area under the curve (AUC) of the receiver operating characteristic curve and Moran's I index were used to assess the performance of the model. The experimental results showed that SCBR-CA yields slightly better AUC and Moran's I index values than CNN-CA, and the dynamically predicted landslide hazard zoning results are equivalent or superior to those of static zoning, which indicates that the SCBR-CA model effectively predict local landslide hazards. [ABSTRACT FROM AUTHOR]

Vision Transformers (ViTs) have become one of the main architectures in deep learning with the self-attention mechanism, and are becoming an alternative to Convolutional Neural Networks (CNNs) for remote sensing scene classification tasks. However, the earlier self-attention layer of ViTs focuses on local features rather than global features, and the deeper self-attention layer focuses on global features but ignores the different impact of different frequency information. This will greatly increase the training and computational cost due to the quadratic complexity of the self-attention mechanism on the long sequence representation. In this paper, we propose a hybrid CNN – vision transformer structure (HCVNet), which uses convolutional layers to replace the earlier self-attention layers, and a novel Frequency Multi-head Self Attention (F-MSA) mechanism to replace the deeper self-attention layers. Specifically, F-MSA is a dual-stream structure that reduces the computational cost and improves the classification performance by encoding the high/low frequency information separately. In addition, a Semantic-aware Localization (SaL) module is introduced, which can guide the selection of crop by learning prior knowledge, avoiding the issue of pure background sampling. Our method performed an accuracy value of 97.20 ± 0.02% on the Aerial Image Dataset and 93.89 ± 0.03% on the NWPU-RESISC45 Dataset, with low complexity costs. [ABSTRACT FROM AUTHOR]

A pulse-like ground motion identification method is proposed based on the Siamese Convolutional Neural Networks (SCNNs). The wavelet coefficient graphs of pulse-like ground motions are used as the input data, and the features extracted by the SCNNs are used for the identification. Based on the time-domain features, pulse-like and non-pulse-like ground motions are classified into several classes. The results indicated that all the identified pulse-like ground motions have similar pulse features to the pre-selected training data. The principal advantage of the method is that the misclassification problem can be minimized by an updating procedure proposed by this paper. [ABSTRACT FROM AUTHOR]

High-performance and low-cost emission characteristic prediction is very crucial for diesel engine design optimization and emission aftertreatment control and diagnosis. In this paper, a novel hybrid model that combines Convolutional Neural Network (CNN) and Gated Recurrent Unit Network (GRU) was proposed to predict the emission characteristic from diesel engines, encompassing CO, THC, CO2, NOx, exhaust temperature and exhaust pressure. Nine operating parameters from WHTC and WHSC cycles, including speed, torque, intake pressure, intake flow, intake temperature, oil pressure, fuel rate, oil temperature, water temperature, were considered as inputs. Firstly, the importance of each variable is evaluated by Random Forests algorithm to determine the optimal inputs for each emission characteristic parameter and reduce redundancy. Then the effect of different hyperparameters on the model performance was discussed in detail and PSO algorithm was used to obtain the optimal hyperparameters. Finally, the CNN-GRU hybrid model was assessed for its generalization and compared with ANN, LSTM and GRU models. The result demonstrates that the CNN-GRU hybrid model with PSO optimization has excellent prediction performance in either the training dataset or the validation dataset. The average value of R² is 0.993 in the training dataset and 0.985 on the validation dataset. In the test dataset, the average R² is 0.961, showing a minor decrease of 3.19% and 2.47% compared to the training and validation dataset, respectively. This indicates that the CNN-GRU hybrid model has strong generalization ability. Compared with other algorithms in the test dataset, the CNN-GRU hybrid model exhibits better comprehensive performance, with the average R² value exceeding that of ANN, LSTM and only GRU by 5.96%, 2.69% and 3.23%, respectively. [ABSTRACT FROM AUTHOR]

Hyperspectral Unmixing (HU), which involves decomposing mixed pixels in hyperspectral images into endmembers and abundance fractions, plays a critical role in spectral analysis. Recent advancements in deep learning methods have shown successful applications in hyperspectral analysis. This paper proposes an end-to-end network structure that combines a Convolutional Neural Network (CNN) and a Multilayer Perceptron (MLP). The 3D-CNN extracts high-level features from both spatial and spectral neighbourhoods of hyperspectral image patches, while the MLP maps these abstracted features to abundance fractions. Considering the challenge of acquiring manual labels, we adopt a Semi-Supervised learning approach that leverages unlabelled data to enhance training. We divide our training data into labelled and unlabelled subsets, defining distinct loss functions for each. Additionally, we introduce a neighbourhood similarity constraint to fully exploit spatial context information. To address scenarios where no labelled data is available, we employ Generative Adversarial Networks (GANs) to generate synthetic labelled samples. Experimental evaluations conducted on both synthetic and real datasets validate the effectiveness of our proposed method, which outperforms several commonly used and state-of-the-art HU methods. [ABSTRACT FROM AUTHOR]

This document is a preface for a special issue of the journal "Nondestructive Testing & Evaluation" focused on non-destructive testing (NDT) and structural health monitoring (SHM) techniques in the context of smart cities. The authors emphasize the importance of NDT and SHM in developing sustainable and resilient urban infrastructure within smart cities. The special issue includes 11 papers that cover a range of topics, including wireless sensing, feature extraction, visual attention, and artificial intelligence (AI). The papers explore applications in urban infrastructure, traffic dynamics, data processing and analysis, and the use of AI for NDT and SHM. The authors express gratitude to the journal for the opportunity to compile these articles and provide readers with insights into advanced techniques and ongoing solutions in NDT and SHM. [Extracted from the article]