Search

Accurate power load prediction is an important guide for power system planning and operation. High‐ or low‐load prediction results will affect the operation of the power system. In recent years, deep learning technology represented by convolution neural network (CNN) and transformer has been proved to be suitable for power load prediction. This paper proposes a new short‐term power load hybrid forecasting model, called channel enhanced attention (CEA) and temporal convolutional network (TCN)‐based transformer comprehensive forecasting model. This method combines the short‐term feature extraction ability of TCN with the long‐term dependent capture ability of transformer for short‐term load forecasting. And the CEA designed in this study is added to improve the prediction accuracy. On the same dataset, the designed model predicts power load mean square errors of 0.056 and 0.146 for the next 24 h and the next week, respectively, which is 0.002 to 0.073 and 0.012 to 0.024 lower than the baseline model. The experimental results show that the hybrid short‐term power load prediction model proposed in this paper is significantly better than the existing methods. The predicted curve is in agreement with the actual charge change, which provides a good guidance for short‐term power load prediction. [ABSTRACT FROM AUTHOR]

The aptness of machine learning (ML) to learn from large datasets, discover trends, and make predictions has demonstrated its potential to metamorphose the medical field. Medical data analysis with ML algorithms can improve patient outcomes in terms of both treatment and diagnosis. This paper investigates the numerous possibilities of ML in the medical industries, including radiology, pathology, genomics, and clinical decision‐making. It also goes over the benefits and drawbacks of ML in various sectors as well as the limitations that come with its application. It illustrates the potential advantages of ML, such as better accuracy and efficiency in diagnosis and individualized treatment programs, through a review of previous studies. Lastly, it provides perspectives on prospective advancements and prospects for the discipline. This study also intends to investigate the applications of deep learning (DL) in the medical field. DL algorithms have performed exceptionally satisfactorily in several healthcare‐related fields. The main conclusions of the study are summarized, and their ramifications for the healthcare sector are discussed in this paper's conclusion. This paper intends to contribute to a greater understanding of the prevailing state of the discipline and the possibility for future developments by emphasizing the prospects of these methodologies to alter medical study and clinical practice. [ABSTRACT FROM AUTHOR]

This article discusses a special issue of the International Conference on Computing and Communications Networks (ICCCN 2022), which focuses on various topics in computing and networking. The special issue includes 12 selected papers that cover a range of subjects, such as code-mixed language identification, sentiment analysis, sleep quality assessment, gait analysis, cervical precancerous lesion classification, marine radar monitoring, text summarization, software optimization, fault diagnosis, imbalanced data analysis, neonatal hypoglycemia brain injury, and feature selection for classification problems. The authors present their research methodologies, experimental results, and conclusions for each paper. The guest editors hope that the research contributions in this special issue will be beneficial to readers. [Extracted from the article]

The surface texture of asphalt pavement has a significant effect on skid resistance performance. However, its contribution to the performance of skid resistance is non‐homogeneous and subjects to local validity. There are also a few deep learning models that take into account the effective contact texture region. This paper proposes a convolutional neural network model based on the effective contact texture region, containing macro‐ and micro‐scale awareness sub‐modules. In this study, the asphalt mixture with varying gradations was designed to accurately obtain the effective contact texture region. Then, the textures were disentangled into macro‐ and micro‐texture scales by applying the fast Fourier transform and fed into the model for training. Finally, the area of effective contact texture region was calculated, and the effective contact ratio parameter was then proposed using the triangulation algorithm. The results showed that the effective contact texture area of pavement varies by the asphalt mixture type. The effective contact ratio parameter exhibited a significant positive correlation (Pearson correlation coefficient is 0.901, R2= 0.8129) with skid resistance performance and was also influenced by key sieve aggregate content from 2.36 to 4.75 mm. The data of effective contact texture region following disentanglement significantly released the model performance (the relative error dropped to 1.81%). The model exhibited improved precision and performance, which can be utilized as an efficient, non‐contact alternative method for skid resistance analysis. [ABSTRACT FROM AUTHOR]

The most recent and precise biological and healthcare knowledge is critical in the current outbreak such as COVID. In today's small world, everyone needs timely and appropriate medical information to prevent contagious diseases. Extraction of important information from medical conversations and dissemination to patients and doctors may benefit in the treatments of doctor tiredness and patient amnesia. Problem: Automatic text summarizing is essential for gaining great knowledge in any topic in an efficient and productive manner. The material included in health records is vital to our understanding of kind illness and its manifestations. Creating a comprehensive and standard kind of content is becoming an unavoidable and crucial problem in the medical process as a result of the massive amounts of fragmented data created in many sectors. Approach: The purpose of this study is to employ NLP‐based deep learning algorithms for text summary that perform well on linguistic text summarization data, and then modify/adapt these for biomedical domain‐specific text summarization. This paper provides an approach developed in‐house for condensing ill‐punctuated or unpunctuated discussion transcripts into more intelligible summaries, which combines topic modelling and phrase selection with punctuation restorations. For autonomous synthesis of medical reports from biomedical transcripts, this research proposes using an end‐to‐end summarization technique, Deep Dense Long Short Term Memory Network (LSTM), followed by Convolutional Neural Network (CNN). Results: Extensive testing, examination, and comparing have demonstrated that this summarizer works well for medical transcript summarization. The suggested approach achieved an average ROUGE score of 93.5% using a single document summary. Furthermore, by comparing new techniques to previous ones, the utility and accuracy of novel strategies would be shown. The results reveal that models trained on ordinary language provide comparable results on a biomedical testing set, with one model outperforming the linguistic test set. [ABSTRACT FROM AUTHOR]

Whole brain segmentation from magnetic resonance images (MRI) is crucial in diagnosing brain diseases and analyzing neuroimaging data. Despite advances through deep learning, challenges such as uneven gray distribution and the presence of artifacts still present hurdles in medical image processing. These limitations are often a result of insufficient spatial contextual information and lack of attention to important regions within existing models. To address these issues, this paper presents SSA‐UNet (Squeeze‐and‐Excitation and Self‐Attention UNet), a uniquely designed deep convolutional neural network that integrates spatial constraints by converting three consecutive 2D MRI slices into a single 2.5D image. This facilitates capturing inter‐slice dependencies effectively. Additionally, the newly formulated SSA block, which sequentially incorporates channel attention and Self‐Attention mechanisms, is placed before the decoders in the conventional U‐Net architecture. This enables the network to automatically weight different feature maps and focus more effectively on regions requiring precise segmentation. Rigorous evaluations on LPBA40 and IBSR18 datasets substantiate the remarkable improvements in accuracy and stability achieved by SSA‐UNet. Results indicate Dice coefficients of 98.38% and 97.47%, specificity of 99.69% and 99.57%, and sensitivity of 98.5% and 97.98% for the respective datasets. Compared to other existing models, SSA‐UNet shows significant improvements on both the LPBA40 and IBSR18 datasets. On the LPBA40 dataset, SSA‐UNet's Dice coefficient improved by 0.33% compared to the sub‐optimal model, while on the IBSR18 dataset, the improvement reached 1.78%. These empirical findings demonstrate SSA‐UNet's heightened capability in addressing the long‐standing challenges in MRI‐based whole‐brain segmentation. [ABSTRACT FROM AUTHOR]

The problems of insufficient recognition accuracy, poor real‐time performance and lack of consideration of actual working conditions in the process of intelligent construction of coal mines make this technology still in the research stage and not applied in practical engineering. The purpose of this paper is to establish an accurate and real‐time recognition model, which can quickly distinguish the vibration acceleration signals of coal and gangue under the influence of external factors such as impact position, velocity, and direction by using the different physical properties of coal and gangue particles. Therefore, the accuracy and real‐time of coal gangue recognition model established by different convolutional neural networks (CNN) structures and different position signal input are studied. First, to meet the real‐time requirements, an original CNN recognition model composed of single convolution layer and single pooling layer is established, and the data collected by seven sensors are input in the form of two‐dimensional matrix. However, the stability of the training and test results is insufficient. To solve this problem, once improved CNN (OI‐CNN) recognition model with multiconvolution layers and multipooling layers is built by deepening the network. The experimental results show that the stability and accuracy are improved, but the real‐time performance is poor. Furthermore, through parameter adjustment, the OI‐CNN is changed to the twice improved CNN (TI‐CNN), and the sensor data at different positions are input in the form of one‐dimensional vectors. The results show that the accuracy and real‐time performance of the TI‐CNN coal gangue recognition model are further improved. Finally, according to the research purpose of this paper, the weights of CNN indexes are given, and a multi‐index comprehensive evaluation system (MICES) is established. With the original CNN recognition model as the control, the OI‐CNN recognition model and the TI‐CNN recognition model at different positions are quantitatively compared to obtain the comprehensive evaluation scores of each model. The results show that the MICES of the coal gangue recognition model established based on the TI‐CNN structure and the data input of a single position sensor is the highest, while the sensor position has little effect on the recognition results. [ABSTRACT FROM AUTHOR]

Text-independent speaker verification (TI-SV) is a crucial task in speaker recognition, as it involves verifying an individual's claimed identity from speech of arbitrary content without any human intervention. The target for TI-SV is to design a discriminative network to learn deep speaker embedding for speaker idiosyncrasy. In this paper, we propose a deep speaker embedding learning approach of a hybrid deep neural network (DNN) for TI-SV in FM broadcasting. Not only acoustic features are utilized, but also phoneme features are introduced as prior knowledge to collectively learn deep speaker embedding. The hybrid DNN consists of a convolutional neural network architecture for generating acoustic features and a multilayer perceptron architecture for extracting phoneme features sequentially, which represent significant pronunciation attributes. The extracted acoustic and phoneme features are concatenated to form deep embedding descriptors for speaker identity. The hybrid DNN demonstrates not only the complementarity between acoustic and phoneme features but also the temporality of phoneme features in a sequence. Our experiments show that the hybrid DNN outperforms existing methods and delivers a remarkable performance in FM broadcasting TI-SV. [ABSTRACT FROM AUTHOR]

This paper addressed the challenges associated with the complexity, numerous parameters, computational resource demands, and slow processing speed of transformer fault identification models based on deep learning technologies. Sparse convolutional neural network (CNN) approach is proposed for identifying faults related to dissolved gases in oil. Leveraging an improved Gramian angular field, one‐dimensional fault samples are converted into two‐dimensional feature images and data augmentation is implemented to meet the input requirements of deep learning models. Building upon visual geometry group (VGG)19 and residual networks (ResNet)50 networks for fault diagnosis, sparsity techniques are introduced through pruning, the fusion of convolution layers and batch normalization layers, and parameter quantization. Numerical experiments and performance evaluations on dissolved gas in transformer oil fault data demonstrate that the proposed method effectively reduced model complexity, minimized parameter count, conserved computational resources, and improved processing speed while maintaining a considerable level of fault identification accuracy. This made it applicable to edge computing platforms characterized by small form factors and low power consumption in the power industry. [ABSTRACT FROM AUTHOR]

This study proposes a novel method, which is a U‐shaped convolutional neural network that combines non‐local attention mechanisms, Res2net residual modules, and terrain information (UNR‐Net). The original U‐Net method and the linear regression (LR) method are conducted as benchmarks. Generally, the UNR‐Net has demonstrated promise in performing a 10× downscaling for daily 2‐m temperature over North China with lead times of 1–7 days and shows superiority to the U‐Net and LR methods. To be specific, U‐Net and UNR‐Net demonstrate higher Nash‐Sutcliffe Efficiency coefficient values compared to LR by 0.052 and 0.077, respectively. The corresponding improvements in pattern correlation coefficient are 0.013 and 0.016, while the root mean square error values are higher by 0.22 and 0.338, respectively. Additionally, the structural similarity index metric is higher by 0.033 and lower by 0.015. Furthermore, regions with significant errors are primarily distributed in complex terrain areas such as the Taihang Mountains, where UNR‐Net exhibits noticeable improvements. In addition, the 12 components‐based error decomposition method is proposed to analyze the error source of different models. Generally, the smallest errors are observed during the summer season and the sequence error component is proven to be the main source error of 2‐m temperature forecasts. Furthermore, UNR‐Net consistently demonstrates the lowest errors among all 12 error components. Therefore, combining the numerical weather prediction model and deep learning method is very promising in downscaling temperature forecasts and can be applied to routine forecasting of other atmospheric variables in the future. Plain Language Summary: This research proposes a new method for downscaling using deep learning. The method uses a specific type of neural network called UNR‐Net, which combines attention mechanisms, residual modules, and terrain information. The performance of UNR‐Net is compared to two other methods: U‐Net and LR. In the study, UNR‐Net shows promise in performing a 10× downscaling of the daily 2‐m temperature in North China. The UNR‐Net demonstrates the best overall performance among all the comprehensive indicators (NSE, pattern correlation coefficient, root mean square error, and structural similarity index metric). Errors in the predictions are mainly found in complex terrain areas like the Taihang Mountains, but UNR‐Net shows noticeable improvements in these regions. The study also proposes a 12 components‐based error decomposition method to analyze the error sources of different models. All in all, it is found that the smallest errors are observed during the summer season and the main source error is the sequence error component. Additionally, when considering lead times of 1–7 days, UNR‐Net consistently shows the lowest errors among all 12 error components. Based on these findings, combining numerical weather prediction models with deep learning methods holds great promise for generating high‐resolution temperature forecasts. Key Points: This paper presents a novel deep learning downscaling method, UNR‐Net, capable of downscaling daily 2‐m temperature by a factor of 10The overall performance of the UNR‐Net method surpasses the U‐Net method and linear regression methodThe 12 components‐based error decomposition method is proposed to analyze the error source of different models [ABSTRACT FROM AUTHOR]

This paper presents a multi‐input deep learning‐based joint pilot decontamination and symbol detection (SD) technique for 5G massive multiple‐input multiple‐output (MAMIMO) systems. It consists of a fully convolutional neural network (FCNN) that finds the 5G channel coefficients using pre‐known sounding reference signals (SRSs) under pilot contamination and a symbol detector derived from projected gradient descent iterations. The study considers pilot contamination caused by inter‐cell interferences between SRSs during channel estimation (CE). The proposed scheme accepts entire 5G orthogonal frequency modulated (OFDM) data and least square estimates and produces the transmitted OFDM signal. Simulation experiments demonstrated that the proposed technique has better CE and SD performance with reduced trainable parameters. Moreover, it is faster due to the lowest elapsed time during end‐to‐end OFDM symbol detection. This paper proposes a joint pilot decontamination and signal detection for 5G MAMIMO systems. It achieves better detection performance with the lowest number of trainable parameters and memory requirements. It is applicable in 5G OFDM symbol detection and channel estimation under pilot contamination. [ABSTRACT FROM AUTHOR]

The segmentation of dermoscopic‐based skin lesion images is considered to be challenging owing to various factors. Some of the most tangible reasons include poor contrast near the affected skin lesion, the fuzzy and unpredictable lesion limits, the presence of variations in noise, and capturing images under different conditions. This paper aims to develop an efficient segmentation model for dermoscopic images of different skin lesions based on deep learning. This paper proposes the 11‐layer convolutional deep neural network with two segmentation models trained from start to finish and do not depend on any previous information about the data. The viability, efficiency, and speculation ability of the models are evaluated on the ISIC2018 database. The proposed model achieves 0.903 accuracy and 0.820 Jaccard index in the segmentation of skin lesions. The model shows better performance compared to other image segmentation techniques from the leaderboards of ISIC2018 using deep learning. [ABSTRACT FROM AUTHOR]

In this paper, a novel model with a parallel structure is proposed to predict NOx emissions from coal‐fired boilers by using historical operational data, coal properties, and convolutional neural networks. The model inputs are processed and passed into three parallel subnetworks with well‐designed building blocks. The features learned by the three subnetworks are fused and used to predict NOx emissions from a 330‐MW pulverized coal‐fired utility boiler. A comprehensive comparison of different prediction models based on deep learning algorithms shows that the prediction model proposed in this paper outperforms other prediction models in terms of root mean square error criteria. The results show that the parallel structure is key to obtaining accurate predictions while reducing model complexity. This suggests that the model's performance can be improved by designing the model architecture. [ABSTRACT FROM AUTHOR]

Lightning often causes death, injury, and damage to various facilities and equipment. Accurately detecting the spatial location of lightning occurrence by predicting thunderstorms and lightning is of great significance. Traditional lightning detection systems detect lightning by measuring the sound, light, and electromagnetic field information radiated by lightning. These methods typically have two problems. First, the detection process of lightning signals is susceptible to electromagnetic interference. Second, the equipment cost is high and is not friendly to some lightning detection tasks only targeted at specific scenarios. In order to detect lightning more conveniently, we propose a lightning detection model based on deep learning networks. With the increase in the use of cameras in modern society, designing lightning object detection networks based on deep learning is possible. However, two problems have been found in existing practice: (1) When strong lightning meteorological phenomena occur, the lightning features in the image are covered by bright electric lights, and convolutional neural networks cannot distinguish between strong lightning scenes and strong ultraviolet scenes. (2) The performance of convolutional neural networks is often related to the model's size. The larger the model, the stronger the performance of the network. However, in practical application scenarios, computing resources are insufficient to use sufficiently large networks. In this paper, we propose a simple and effective lightning object detection network (LD‐Net) and use a foreground‐background segmentation algorithm to locate frames containing lightning in the video. After using the knowledge distillation‐based model compression method, the mAP of the lightning object detection network with a backbone net of resnet with 18‐layer (LD‐Net‐18) can reach 82.4%. We hope that the proposed LD‐Net can serve as a simple and powerful alternative to traditional lightning detection methods, enhancing efficiency in lightning detection tasks. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural network (CNN) has shown great power in single image super resolution (SISR) reconstruction, and achieving significant improvements over traditional methods. Despite the great success of these CNN‐based methods, direct application of these methods to some edge devices is impractical due to the large computational overhead required. To address this problem, a novel, lightweight SISR network focusing on speed and accuracy, called the multi‐path feedback fusion network (MFFN), has been designed in this paper. Specifically, in order to extract features more effectively, the authors propose a novel fusion attention feedback block (FAFB) as the main building block of MFFN. The FAFB consists of a backbone branch and several hierarchical branches. The backbone branch is composed of stacked enhanced pixel attentional blocks (EPAB), which are responsible for incremental deep feature learning on the feature map. And the hierarchical branches are responsible for extracting feature maps with different sizes of receptive fields and fusing these feature maps with the features extracted from the trunk branches to achieve multi‐scale feature learning, which the authors refer to this design as the multi‐scale fusion block (MSFB). Extra enhancement information (EIE) is added to each EPAB input, which enables the backbone branch to learn more effectively. On the other hand, the outputs of the cascade branches are further complemented by an additional feedback fusion enhancement block (FFEB) before being fused with the output of the trunk branches to achieve more comprehensive and accurate feature learning. Numerous experiments have shown that MFFN achieves higher accuracy than other state‐of‐the‐art methods on benchmark test sets. [ABSTRACT FROM AUTHOR]

The accurate classification of power quality disturbances (PQDs) is crucial for advancing real‐time monitoring and classification systems within the modern power grid. The proposed system must ensure dependable, safeguarded, and stable operating conditions amidst diverse power quality issues. This paper presents an approach to classifying power quality disturbances using a deep learning model that synergizes deep convolutional neural networks (DCNN) and Bidirectional Long Short‐Term Memory (BiLSTM). This amalgamation effectively extracts and classifies disturbance signals in real time, grounded on noise levels. The initial feature extraction from the signal is accomplished through a time‐frequency matrix. Subsequently, secondary extraction employs the BiLSTM layer to intricately and significantly classify disturbances in the power signal. This aids in transforming high‐dimensional matrices into a reduced set for enhanced performance. The detailed classification is facilitated by the softmax layer. The simulation results support the power quality evaluations under varied constraints and underscore the substantial classification of power quality disturbances through the DCNN‐BiLSTM algorithm, in comparison to alternative classification algorithms in terms of computational speed and accuracy. [ABSTRACT FROM AUTHOR]

Lightweight deep learning models for face recognition are becoming increasingly crucial for deployment on resource‐constrained devices such as embedded systems or mobile devices. This paper presents a highly efficient and compact deep learning (DL) model that achieves state‐of‐the‐art performance on various face recognition benchmarks. The developed DL model employs one‐ or few‐shot learning to obtain effective feature embeddings and draws inspiration from FaceNet with significant refinements to achieve a memory size of only 3.5 MB—about 30 times smaller than FaceNet—while maintaining high accuracy and real‐time performance. The study demonstrates the model's effectiveness through extensive experiments, which include testing on public datasets and the model's ability to recognize occluded faces in uncontrolled environments using grayscale input images. Compared to the state‐of‐the‐art lightweight models, the proposed model requires fewer FLOPs (0.06G), has a smaller number of parameters (1.2 M), and occupies a smaller model size (3.5 MB) while achieving a competitive level of recognition accuracy and real‐time performance. The results show that the model is well‐suited for deployment in embedded domains, including live entrance security checks, driver authorization, and in‐class attendance systems. The entire code of FN8 is available onGitHub. [ABSTRACT FROM AUTHOR]

Introduction: Particularly within the Internet of Medical Things (IoMT) context, skin lesion analysis is critical for precise diagnosis. To improve the accuracy and efficiency of skin lesion analysis, CAD systems play a crucial role. To segment and classify skin lesions from dermoscopy images, this study focuses on using hybrid deep learning techniques. Method: This research uses a hybrid deep learning model that combines two cutting‐edge approaches: Mask Region‐based Convolutional Neural Network (MRCNN) for semantic segmentation and ResNet50 for lesion detection. To pinpoint the precise location of a skin lesion, the MRCNN is used for border delineation. We amass a huge, annotated collection of dermoscopy images for thorough model training. The hybrid deep learning model to capture subtle representations of the images is trained from start to finish using this dataset. Results: The experimental results using dermoscopy images show that the suggested hybrid method outperforms the current state‐of‐the‐art methods. The model's capacity to segment lesions into distinct groups is demonstrated by a segmentation accuracy measurement of 95.49 percent. In addition, the classification of skin lesions shows great accuracy and dependability, which is a notable advancement over traditional methods. The model is put through its paces on the ISIC 2020 Challenge dataset, scoring a perfect 96.75% accuracy. Compared to current best practices in IoMT, segmentation and classification models perform exceptionally well. Conclusion: In conclusion, this paper's hybrid deep learning strategy is highly effective in skin lesion segmentation and classification. The results show that the model has the potential to improve diagnostic accuracy in the setting of IoMT, and it outperforms the current gold standards. The excellent results obtained on the ISIC 2020 Challenge dataset further confirm the viability and superiority of the suggested methodology for skin lesion analysis. [ABSTRACT FROM AUTHOR]

Automatic classification of Lyme disease rashes on the skin helps clinicians and dermatologists' probe and investigate Lyme skin rashes effectively. This paper proposes a new in‐depth features fusion system to classify Lyme disease rashes. The proposed method consists of two main steps. First, three different deep learning models, Densenet201, InceptionV3, and Exception, were trained independently to extract the deep features from the erythema migrans (EM) images. Second, a deep feature fusion mechanism (meta classifier) is developed to integrate the deep features before the final classification output layer. The meta classifier is a basic deep convolutional neural network trained on original images and features extracted from base level three deep learning models. In the feature fusion mechanism, the last three layers of base models are dropped out and connected to the meta classifier. The proposed deep feature fusion method significantly improved the classification process, where the classification accuracy was 98.97%, which is particularly impressive than the other state‐of‐the‐art models. [ABSTRACT FROM AUTHOR]

Depression is a severe mental health issue. The user‐generated content on social media (SM) is growing nowadays. Some computational approaches have been proposed for detecting depression based on users' SM data. However, because of the use of formal language, short range of words and misspellings in the SM data, depression detection (DD) is a challenging task. This paper proposes a novel deep learning (DL) technique for performing DD of the SM data with the help of the hybrid feature selection (FS) mechanism. Initially, two publicly available datasets containing user tweets are collected for implementing the proposed research model. Then the collected datasets are preprocessed for further processing. The preprocessing phase includes critical processes that contribute to creating a ready‐to‐use dataset for training and testing. After preprocessing, the preprocessed data is divided into prime and non‐prime words based on the dictionary approach. After that, the hybrid FS approach is implemented to select the most relevant features from the prime and non‐prime words for higher classification accuracy (AC). In the hybrid model, firstly Term Frequency Inverse Document Frequency integrated Modified Information Gain (TFIDF‐MIG) approach is proposed that assigns the score value of each prime and non‐prime word in the dataset. Secondly, optimal features are selected from the weighted features using the Improved Elephant Herding Algorithm (IEHA). Finally, the decided features from the hybrid model are fed into the DL model, namely attention included improved ReLU‐based Convolution Neural Network with Long Short‐Term Memory (AIRCNN‐LSTM) for DD. Experiments are performed on the collected datasets to assess the proposed model's performance efficiency. The results of the extensive experiments show that the presented work outperforms existing techniques regarding DD classification AC by locating the best solutions. At the same time, it reduces the number of features chosen. [ABSTRACT FROM AUTHOR]

Text regression is an important task in natural language processing (NLP), which aims to predict continuous numerical values associated with text. Previous work focused on linear text regression requiring manual feature selection for financial prediction. Recently, non‐linear text regression through neural network models has become a trend. However, most models rely only on convolutional neural networks (CNN) and suffer from insufficient interpretability. In this paper, we propose a deep neural network model named EM‐CBA for text regression and further interpret the model. The proposed model is powered by word EMbedding, CNN, Bidirectional long short‐term memory (Bi‐LSTM) and Attention mechanism. The proposed EM‐CBA takes financial report texts as input and predicts a financial metric named return on assets (ROA). We conduct comprehensive experiments on a dataset about the reports of enterprises. Experimental results show that the proposed model provides more accurate predictions of enterprises' metrics than previous convolutional neural network models and other classical models. The validity of each module of the model is also verified. Finally, we demonstrate a way of performing analysis in words change and results errors to intuitively interpret the effect of different text inputs on the model. The analysis demonstrates that the model is able to use information about sentiment words to analyse their associated contexts to revise the predictions. [ABSTRACT FROM AUTHOR]

In recent years, deep learning‐based fault diagnosis technology for high‐voltage circuit breakers (HVCB) has advanced significantly, but the working environment of HVCBs is complex, resulting in unsatisfactory fault diagnosis results of HVCBs in noisy environment and existing deep learning methods are difficult to solve this problem. This paper proposes a multi‐channel convolutional neural network combines dense residual structure and attention mechanism to achieve high‐precision and high‐robust diagnosis of HVCBs in noisy backgrounds. A dense residual network is introduced into the convolutional neural network to prevent feature loss during network propagation to preserve the difference information between the network layers as much as possible, Simultaneously, a channel attention mechanism is introduced to adaptively adjust the weights of different convolution channels. The model can extract multi‐scale features from the original signal and fully exploit the intrinsic relationship between the vibration signal and the HVCB's operating state. The experimental results show that the diagnostic method can still meet the requirements of fault diagnosis in the presence of noise, with an average diagnostic accuracy rate of 85.92% when the signal‐to‐noise ratio is −4. The model outperforms the traditional single‐channel model in terms of diagnostic accuracy and stability. [ABSTRACT FROM AUTHOR]

Alzheimer's disorder (AD) causes permanent impairment in the brain's memory of the cellular system, leading to the initiation of dementia. Earlier detection of Alzheimer's disease in the initial stages is challenging for researchers. Deep learning and machine learning‐based techniques can help resolve many issues associated with brain imaging exploration. Brain MR Images (Brain‐MRI) are used to detect Alzheimer's in computable research work. To correctly categorize the stages of Alzheimer's disease, discriminative features need to be extracted from the MR images. Recently, many studies have used deep learning methods for the early detection of this disorder. However, overfitting degrades the deep learning method's performance because the dataset's selection images are smaller and imbalanced. Some studies could not reach more discriminative and effectual attention‐aware features for Alzheimer's stage classification to increase the model performance. In this paper, we develop a novel hierarchical residual attention learning‐inspired multistage conjoined twin network (HRAL‐CTNN) to classify the stages of Alzheimer's. We used augmentation approaches to scale insufficient and imbalanced data. The HRAL‐CTNN is efficiently overcoming the issues of not obtaining efficient attention‐aware and generative features for Alzheimer's stage classification. The proposed model solved the problem of redundant features by extracting attentive discriminant features, and scaling imbalance data by data augmentation, after that training and validation using HRAL‐CTNN. The execution of this proposed work has been performed on the ADNI MRI dataset. This work achieved outstanding accuracy of 99.97 ±$$ \pm $$ 0.01% and F1 score of 99.30 ±$$ \pm $$ 0.02% for Alzheimer's stage classification. This model proposed by our group outperformed the existing related studies in terms of the model's performance score. [ABSTRACT FROM AUTHOR]

With the surge of searching and reading online health‐based articles, maintaining the quality and credibility of online health‐based articles has become crucial. The circulation of deceptive health information on numerous social media sites can mislead people and can potentially cause adverse effects on people's health. To address these problems, this work uses deep learning approaches to automate the assessment and scoring of online health‐related articles' credibility. The paper proposed an Attention‐based Recurrent Multichannel Convolutional Neural Network (ARMCNN) model. The proposed model incorporates a BiLSTM layer, a multichannel CNN layer, and an attention layer and predicts the credibility of online health information. To perform a reliable evaluation of the presented model, we utilize the health articles reviewed by the experts, compiled in a labeled dataset termed "Pubhealth," which consists of thousands of health articles. The results are evaluated using five performance measures, accuracy, precision, recall, f1‐score, and area under the ROC curve (AUC). Furthermore, we extensively compared the proposed model with different deep learning and machine learning models such as Long short‐term memory (LSTM), Bidirectional LSTM, CNN (Convolutional neural network), and RNN‐CNN. The experimental results showed that the proposed model produced state‐of‐the‐art performance on the used dataset by achieving an accuracy of 0.88, precision of 0.92, recall of 0.87, f1‐score of 0.90, and AUC of 0.94. Further, the proposed model yielded better performance than other benchmarked techniques for the credibility assessment of online health articles. [ABSTRACT FROM AUTHOR]

It has been widely recognized that tropical cyclone (TC) genesis requires favorable large‐scale environmental conditions. Based on these linkages, numerous efforts have been made to establish an empirical relationship between seasonal TC activities and large‐scale environmental favorability in a quantitative way, which lead to conceptual functions such as the TC genesis index. However, due to the limited amount of reliable TC observations and complexity of the climate system, a simple analytic function may not be an accurate portrait of the empirical relationship between TCs and their ambiences. In this research, we use convolution neural networks (CNNs) to disentangle this complex relationship. To circumvent the limited amount of seasonal TC observation records, we implement transfer‐learning technique to train ensemble of CNNs first on suites of high‐resolution climate model simulations with realistic seasonal TC activities and large‐scale environmental conditions, and then on a state‐of‐the‐art reanalysis from 1950 to 2019. The trained CNNs can well reproduce the historical TC records and yields significant seasonal prediction skills when the large‐scale environmental inputs are provided by operational climate forecasts. Furthermore, by inputting the ensemble CNNs with 20th century reanalysis products and Phase 6 of the Coupled Model Intercomparison Project (CMIP6) simulations, we investigated TC variability and its changes in the past and future climates. Specifically, our ensemble CNNs project a decreasing trend of global mean TC activity in the future warming scenario, which is consistent with our future projections using high‐resolution climate model. Plain Language Summary: Tropical cyclones (TCs) require favorable large‐scale environmental conditions to form. Pioneer studies show that these conducive conditions include warm sea surface temperatures (SSTs), sufficient low‐level vorticities and mid‐level humilities, as well as weak‐to‐moderate vertical wind shears. Several follow‐up studies have focused on improving the empirical linkage between number of TC and environmental conditions and developed sets of TC genesis index based on conventional statistical methods. Although these indices can capture climatology of TC genesis spatial distributions and seasonal variation reasonably well, their representations of interannual TC variability are degraded. With the aim to better represent TC interannual variability and long‐term trend using large‐scale environmental conditions, we trained ensembles of convolution neural networks (CNNs) based on the combination of observations and large sets of high‐resolution dynamical climate simulations. The trained CNNs perform significantly well in capturing observed TC interannual‐to‐multidecadal variability and are broadly applicable to many areas of seasonal TC activities. Using a deep learning technique, this paper introduces a new potential avenue to improve our understanding of TC variability and future changes. Key Points: Ensemble convolutional neural networks (CNNs) are trained to emulate seasonal tropical cyclone (TC) activity using environmental factorsThe trained CNNs can be utilized to study seasonal TC variability, and their changes in the past, current and future climatesSkillful seasonal TC predictions can be made using CNN‐based statistical‐dynamical hybrid framework [ABSTRACT FROM AUTHOR]

The behaviour analysis of farmers is very important to gradually improve the sustainable production capacity of the land and promote the economics reasonable increasement. How to effectively extract video features and identify human behaviour is a current research hotspot. In this paper, the authors propose a convolutional neural network model based on the deformable convolution including a residual module and an attention module. The model fuses and encodes video spatial features on the time axis and extracts semantic information. The authors combine the attention mechanism to extract local and global features in the video, which are used to calculate spatial features and micro‐motion features, respectively. Temporal attention mechanism fuses each micro‐motion feature of video to extract higher‐level semantic features of motion. Next, the authors describe their proposed model based on deep learning. First, the authors introduce a video frame sampling method based on the TRN algorithm. Second, they design a feature extraction and fusion model based on the ResNet model and the attention mechanism that extract local and global features in videos. Finally, the authors classify and identify the extracted video features through the classifier. The authors' approach shows comparable performance on publicly available datasets. [ABSTRACT FROM AUTHOR]

Machine learning provides powerful techniques for several applications, including automated disease diagnosis through medical image classification. Recently, many studies reported that deep learning approaches have demonstrated significant performance and accuracy improvements over shallow learning techniques. The deep learning approaches have been used in many problems related to disease diagnoses, such as thyroid diagnosis, diabetic retinopathy detection, foetal localization, and breast cancer detection. Many deep learning methods have been reported in the recent past that uses medical images from various sources, such as healthcare providers and open data initiatives, and reported significant improvement in terms of precision, recall, and accuracy. This paper proposes a framework incorporating deep convolutional neural networks and an enhanced feature extraction technique for classifying medical data. To show the real‐world usability of the proposed approach, it has been used for the classification of COVID‐19 images from computed tomography scans. The experimental results show that the proposed approach outperformed some of the chosen baselines and obtained an accuracy of 98.91%, comparable with already reported accuracies. [ABSTRACT FROM AUTHOR]

This paper proposes a joint optimization method for the imaging algorithm and sampling scheme of sparse spotlight synthetic aperture radar (SAR) imaging based on deep convolutional neural networks. Traditional compressed sensing‐based sparse SAR imaging has been widely studied. Deep learning and sparse unfolding networks have been introduced into sparse SAR imaging, but most current works focus only on the imaging stage and simply adopt the conventional uniform or random downsampling scheme. Considering that the imaging quality also depends on the sampling pattern besides the imaging algorithm, this paper introduces a learning‐based strategy to jointly optimize the sampling scheme and the imaging network parameters of the reconstruction module. In a deep learning‐based image reconstruction scheme, joint and continuous optimization of the sampling patterns and convolutional neural network parameters is achieved to improve the image quality. Simulation results based on real SAR image data set illustrate the effectiveness and superiority of the proposed framework. [ABSTRACT FROM AUTHOR]

Great progress has been made in deep learning image reconstruction, but there are still problems that need to be improved, such as the extraction of image texture details and the improvement of the overall contour quality of the image, and how to reduce the transmission cost is also the focus of research. This paper proposes a fast bilateral network suitable for both grayscale and color image reconstruction. In the compression network, bilinear interpolation, fully connected layer and convolutional neural network are selected for image compression. The reconstruction network is divided into two parts: the texture path and contour path, the former reconstructs the remaining texture details of the image, and the latter performs deep contour reconstruction. The bilateral complementary residual connection method transfers the texture information to the contour path and improves the contour quality, and the improvement of the contour quality can improve the learning of texture details. Through a large number of data tests, it shows that the network in this paper has achieved comparable or better results in terms of image reconstruction quality, time‐consuming and robustness against noise. It solves the problem of memory consumption for storing a large number of images, and also provides convenience for shortening space and time during image transmission. [ABSTRACT FROM AUTHOR]

With a broad usage of digital imaging data in hospitals, the medical image repository size is growing in a very rapid manner. This creates difficulty in querying and managing these vast databases that in turn leads to the requirement of Content‐Based Medical Image Retrieval (CBMIR) systems. The CBMIR is considered as the major ambiguous and challenging task for reducing the semantic gap among the human queries and images in the datasets having more information content. The main intention of this paper is to enhance medical image retrieval and classification using the improved pattern descriptor and deep learning. For the medical image retrieval, this paper develops the Optimized Local Directional Weber Pattern with a multi‐objective similarity function. This similarity function focuses on the measures like Structural Similarity Index Measure, Peak Signal‐to‐Noise Ratio, Mean Squared Error MSE, and correlation. Likewise, the development of an improved Faster‐Region Convolutional Neural Network (Faster‐RCNN) is adopted in the classification phase. Both the retrieval and classification are enhanced by the Modified Wind Speed‐based Deer Hunting Optimization Algorithm. Considering classification and retrieval tasks, the experiments on benchmark datasets reveal an enhancement from the introduced algorithm when compared over the existing methods. [ABSTRACT FROM AUTHOR]

This main intention of this paper is to adopt a new disease detection model for plant leaves. The proposed model involves several steps such as pre‐processing, leaf segmentation, abnormality segmentation, feature extraction and detection. Image scaling and contrast enhancement are performed during the pre‐processing phase. Once the pre‐processing is done, the segmentation phase starts with leaf segmentation by binary thresholding method and abnormality segmentation by K‐means clustering. Further, the local binary pattern and grey‐level co‐occurrence matrix features are extracted and a dimensionality reduced technique called principle component analysis is determined. As the main novelty, the weighted feature extraction is performed, in which the weight functions are optimized by the self‐adaptive deer hunting optimization (SA‐DHOA). Another contribution of this paper is to implement a hybrid classifier for disease detection. Here, the extracted weighted features are subjected to a support vector machine, and the abnormality segmented image is subjected to convolutional neural network (CNN), which is a deep learning algorithm that can learn the features automatically. Here, the same SA‐DHOA is used to improve the performance of CNN, which is termed as SA‐DHOA‐SCNN. Finally, the performance analysis confirms the maximum success rate of the proposed model over other conventional methods. [ABSTRACT FROM AUTHOR]

Objective: Sleep‐related hypermotor epilepsy (SHE) is a focal epilepsy with seizures occurring mostly during sleep. SHE seizures present different motor characteristics ranging from dystonic posturing to hyperkinetic motor patterns, sometimes associated with affective symptoms and complex behaviors. Disorders of arousal (DOA) are sleep disorders with paroxysmal episodes that may present analogies with SHE seizures. Accurate interpretation of the different SHE patterns and their differentiation from DOA manifestations can be difficult and expensive, and can require highly skilled personnel not always available. Furthermore, it is operator dependent. Methods: Common techniques for human motion analysis, such as wearable sensors (e.g., accelerometers) and motion capture systems, have been considered to overcome these problems. Unfortunately, these systems are cumbersome and they require trained personnel for marker and sensor positioning, limiting their use in the epilepsy domain. To overcome these problems, recently significant effort has been spent in studying automatic methods based on video analysis for the characterization of human motion. Systems based on computer vision and deep learning have been exploited in many fields, but epilepsy has received limited attention. Results: In this paper, we present a pipeline composed of a set of three‐dimensional convolutional neural networks that, starting from video recordings, reached an overall accuracy of 80% in the classification of different SHE semiology patterns and DOA. Significance: The preliminary results obtained in this study highlight that our deep learning pipeline could be used by physicians as a tool to support them in the differential diagnosis of the different patterns of SHE and DOA, and encourage further investigation. [ABSTRACT FROM AUTHOR]

In medical science, imaging is the most effective diagnostic and therapeutic tool. Almost all modalities have transitioned to direct digital capture devices, which have emerged as a major future healthcare option. Three diseases such as Alzheimer's (AD), Haemorrhage (HD), and COVID‐19 have been used in this manuscript for binary classification purposes. Three datasets (AD, HD, and COVID‐19) were used in this research out of which the first two, that is, AD and HD belong to brain Magnetic Resonance Imaging (MRI) and the last one, that is, COVID‐19 belongs to Chest X‐Ray (CXR) All of the diseases listed above cannot be eliminated, but they can be slowed down with early detection and effective medical treatment. This paper proposes an intelligent method for classifying brain (MRI) and CXR images into normal and abnormal classes for the early detection of AD, HD, and COVID‐19 based on an ensemble deep neural network (DNN). In the proposed method, the convolutional neural network (CNN) is used for automatic feature extraction from images and long‐short term memory (LSTM) is used for final classification. Moreover, the Hill‐Climbing Algorithm (HCA) is implemented for finding the best possible value for hyper parameters of CNN and LSTM, such as the filter size of CNN and the number of units of LSTM while fixing the other parameters. The data‐set is pre‐processed (resized, cropped, and noise removed) before feeding the train images to the proposed models for accurate and fast learning. Forty‐five MR images of AD, Sixty MR images of HD, and 600 CXR images of COVID‐19 were used for testing the proposed model 'CNN‐LSTM‐HCA'. The performance of the proposed model is evaluated using six types of statistical assessment metrics such as; Accuracy, Sensitivity, Specificity, F‐measure, ROC, and AUC. The proposed model compared with the other three types of hybrid models such as CNN‐LSTM‐PSO, CNN‐LSTM‐Jaya, and CNN‐LSTM‐GWO and also with state‐of‐art techniques. The overall accuracy of the proposed model received was 98.87%, 85.75%, and 99.1% for COVID‐19, Haemorrhage, and Alzheimer's data sets, respectively. [ABSTRACT FROM AUTHOR]

Automated breast ultrasound (ABUS) imaging system is a practical technique to automatically scan the whole breast. Automatic tumor detection plays a significant role in the clinic. However, training deep convolutional neural networks (CNNs) for tumor detection needs a large quantity of labeled data. It is time‐consuming and expensive to manually annotate tumor positions in ABUS images. In this paper, a novel semi‐supervised learning EfficientDet (SSL‐E) model is proposed for ABUS tumor detection. Our SSL‐E model solves the tumor detection problem from high similarity and serious unbalance between tumors and backgrounds. Considering the image contrast variation and tumor scale variations in ABUS images, color transformation and geometric transformation are employed for data augmentations. Then the consistency between image and its augmented version is developed, thus the robustness of the detector can be improved. Aiming at the problem of serious unbalance between tumors and backgrounds, a novel copy‐paste synthesis strategy is designed, which can generate more tumor samples and enhance tumor diversity. This method is tested on 68 tumor volumes and 68 normal volumes, including 43,248 slices (1683 tumor slices and 41,565 normal slices). It obtains a promising result with sensitivity of 90.2% and false positives per image (FPs/I) at 0.15. [ABSTRACT FROM AUTHOR]

Objective: COVID‐19 is ravaging the world, but traditional reverse transcription‐polymerase reaction (RT‐PCR) tests are time‐consuming and have a high false‐negative rate and lack of medical equipment. Therefore, lung imaging screening methods are proposed to diagnose COVID‐19 due to its fast test speed. Currently, the commonly used convolutional neural network (CNN) model requires a large number of datasets, and the accuracy of the basic capsule network for multiple classification is limital. For this reason, this paper proposes a novel model based on CNN and CapsNet. Methods: The proposed model integrates CNN and CapsNet. And attention mechanism module and multi‐branch lightweight module are applied to enhance performance. Use the contrast adaptive histogram equalization (CLAHE) algorithm to preprocess the image to enhance image contrast. The preprocessed images are input into the network for training, and ReLU was used as the activation function to adjust the parameters to achieve the optimal. Result: The test dataset includes 1200 X‐ray images (400 COVID‐19, 400 viral pneumonia, and 400 normal), and we replace CNN of VGG16, InceptionV3, Xception, Inception‐Resnet‐v2, ResNet50, DenseNet121, and MoblieNetV2 and integrate with CapsNet. Compared with CapsNet, this network improves 6.96%, 7.83%, 9.37%, 10.47%, and 10.38% in accuracy, area under the curve (AUC), recall, and F1 scores, respectively. In the binary classification experiment, compared with CapsNet, the accuracy, AUC, accuracy, recall rate, and F1 score were increased by 5.33%, 5.34%, 2.88%, 8.00%, and 5.56%, respectively. Conclusion: The proposed embedded the advantages of traditional convolutional neural network and capsule network and has a good classification effect on small COVID‐19 X‐ray image dataset. [ABSTRACT FROM AUTHOR]

Objective: Facial expressions have been served as clinical symptoms to convey mental conditions in psychiatry. This paper proposes to recognize patients with schizophrenia (SCZ) using their facial images based on deep learning algorithm, and to investigate objective differences in facial expressions between SCZ patients and healthy controls using deep learning algorithm and statistical analyses. Methods: The study consists of two parts. The first part recruited 106 SCZ patients and 101 healthy controls, and videotaped their facial expressions through a fixed experimental paradigm. The video data were randomly divided into two sets, one for training a convolutional neural network (CNN) with the classification of "healthy control" or "SCZ patient" as output and the other for evaluating the classification result of the trained CNN. In the second part, all facial images of the recruited participants were put into another CNN separately, which was priorly trained with a facial expression database and will output the most likely facial expressions of the recruited participants. Statistical analyses were performed on the obtained facial expressions to find out the objective differences in facial expressions between the two recruited groups. Results: The trained CNN achieved an overall accuracy of 95.18% for classifying "healthy control" or "SCZ patient." Statistical results on the obtained facial expressions demonstrated that the objective differences between the two recruited groups were statistically significant (p <.05). Conclusions: Facial expressions hold great promise as SCZ clues with the help of deep learning algorithm. The proposed approach would be potentially applied to mobile devices for autorecognizing SCZ in the context of clinical and daily life. [ABSTRACT FROM AUTHOR]

Depression is a serious mental health condition that may lead to poor mental and emotional functioning at work, at school and in the family causing the mental imbalance. In worst scenarios, depression may lead to severe anxiety or suicide. Hence, it is necessary to diagnose depression at early stages. This paper elaborates the development of a novel approach for a convolutional neural network model that can examine facial images from the recorded interview sessions to discover facial patterns that could indicate depression level. The user‐generated data helps to distinguish between different depressive groups with depression symptoms that can manifest people with various mental illnesses in different ways. In particular, we want to automatically predict the depression scale and differentiate depression from other mental disorders using the patient's psychiatric illness history and dynamic textual descriptions extracted from the user inputs. We apply the k‐nearest neighbour algorithm on the dynamic textual descriptors to make a linguistic analysis for classifying mental illness into different classes. We apply dimensionality reduction and regression using the Random Forest algorithm to predict the depression scale. The proposed framework is an extension to pre‐existing frameworks, replacing the handcrafted feature extraction technique with the deep feature extraction. The model performs 2.7% better than existing frameworks in facial detection and feature extraction. [ABSTRACT FROM AUTHOR]

Deep learning has been widely used for medical image segmentation and a large number of papers has been presented recording the success of deep learning in the field. A comprehensive thematic survey on medical image segmentation using deep learning techniques is presented. This paper makes two original contributions. Firstly, compared to traditional surveys that directly divide literatures of deep learning on medical image segmentation into many groups and introduce literatures in detail for each group, we classify currently popular literatures according to a multi‐level structure from coarse to fine. Secondly, this paper focuses on supervised and weakly supervised learning approaches, without including unsupervised approaches since they have been introduced in many old surveys and they are not popular currently. For supervised learning approaches, we analyse literatures in three aspects: the selection of backbone networks, the design of network blocks, and the improvement of loss functions. For weakly supervised learning approaches, we investigate literature according to data augmentation, transfer learning, and interactive segmentation, separately. Compared to existing surveys, this survey classifies the literatures very differently from before and is more convenient for readers to understand the relevant rationale and will guide them to think of appropriate improvements in medical image segmentation based on deep learning approaches. [ABSTRACT FROM AUTHOR]

The combination of computer algorithms to diagnose clinical images has attracted more and more attention. This research aims to improve the efficiency of gastric cancer (GC) diagnosis, so deep learning (DL) algorithms are tentatively used to assist doctors in the diagnosis of gastric cancer. In the experiment, the collected 3591 gastroscopic images were divided into network training set and experimental verification test set. The lesion samples in the image are all marked by many endoscopists with many years of clinical experience. In order to improve the experimental effect, 5261 endoscopic images were obtained by expanding the training set. Then the obtained training set is input into the convolutional neural network (CNN) for training, and finally get the algorithm model DLU‐Net. By inputting 598 test set samples into the CNN constructed in this paper, five results such as advanced GC, early GC, precancerous lesions, normal and benign lesions can be identified and output, with a total accuracy of 94.1%. It can be concluded that the DL algorithm model constructed in this paper can effectively identify the staging characteristics of cancer as well as gastroscopic images, greatly improve efficiency, and effectively assist physicians in the diagnosis of GC under gastroscopy. [ABSTRACT FROM AUTHOR]

Different from most current studies using convolutional neural network (CNN), a deep learning detection method for steel plate surface defects based on Transformer is researched. This paper presents a process, network structure and detection method for steel strip surface defect detection. A Swin Transformer was used to extract hierarchical features in the detection system. Then feature pyramid networks (FPN) was used to fuse the above features to form multi‐scale feature maps, and region proposal network (RPN) was adopted to generate the generate region of interest (ROI) of defects. Finally, the ROI head was used to generate defect category information and its precise location. Here, ablation experiments were conducted to explore the impact of different backbone networks and the number of stages of Swin Transformer and FPN on target detection performance to prove the rationality and efficiency of the network. On the NEU‐DET dataset, the proposed algorithm achieved 81.1% mAP, which is higher than that of classic CNN detection methods such as Faster R‐CNN, SSD, Yolo v3 and RepPoints. [ABSTRACT FROM AUTHOR]

In this paper, a novel accurate and economical 3D computer vision‐based framework is proposed to quantify out‐of‐plane displacements of steel plate structures. First, a sequence of image frames of the steel plate structures of interest is collected. Second, using image association, structure‐from‐motion, and multi‐view stereo algorithms, a 3D point cloud of the steel plate structures and their surroundings is created. Third, an efficient 3D object detection method based on convolutional neural networks is developed and implemented to identify the steel plate structures in the 3D point cloud. Last, the out‐of‐plane displacements of the steel plate structures are quantified using point cloud postprocessing algorithms. The proposed framework has been implemented on a steel plate damper and a full‐scale steel corrugated plate wall panel, which are commonly used in structural and earthquake engineering applications. The results indicate the developed framework can successfully localize the steel plate components in the 3D scene and accurately quantify the out‐of‐plane structural displacements with an average accuracy of ∼1 mm. The implementation shows the proposed framework can accurately and efficiently quantify the out‐of‐plane displacements of steel plate structures in realistic engineering applications. [ABSTRACT FROM AUTHOR]

In this paper, a large‐scale dataset called the Iran Autonomous Driving Dataset (IADD) is presented, aiming to improve the generalization capability of the deep networks outside of their training domains. The IADD focuses on 2D object detection and contains more than 97,000 annotated images, covering six common object classes in the field of autonomous vehicles. To improve the generalization of the models, a wide variety of driving conditions and domains, including the city and suburban road settings, adverse weather conditions, and various traffic flows, are presented in the IADD images. The results of exhaustive evaluations conducted on several state‐of‐the‐art convolutional neural networks reveal that not only the trained architectures have performed successfully on test data of the IADD, but also they have upheld high precision in the assessments of generalization capability. In order to challenge the models, broad range of simulations have been performed in the CARLA software environment; which due to the synthetic nature of the simulated images, severe domain shifts have been observed between the CARLA and the IADD. Also, the cross‐domain evaluation results have confirmed the efficacy of the IADD in enhancing the generalization ability of the deep learning models. The dataset is available in: https://github.com/ahv1373/IADD. [ABSTRACT FROM AUTHOR]

Aiming at the issues of low individual identification accuracy of radiation sources under the condition of little samples, this paper proposes a way of individual identification of radiation source based on strengthening global deep convolutional generative adversarial network (SGDCGAN) and deep convolutional neural network (DCNN) to achieve data augmentation. The method first performs IQ map feature splicing processing on the input signal, and then uses DCNN to automatically obtain the deep essential features of the data. Besides, an adaptive improvement is created to the deep convolutional generative adversarial network, and a self‐attention mechanism is introduced into the discriminator and the generator to reinforce the integrity and authenticity of the generated samples. For the common gradient disappearance during model training, the gradient penalty mechanism and spectral normalization are added to make the training process more stable. Through comparison experiments on the collected ADS‐B signals, the experimental results have demonstrated that when the signal‐to‐noise ratio is 0 dB and the number of original samples in each class is 40, the recognition accuracy is improved by 23.6% after doubling the data. Compared with the DCGAN‐DCNN and GAN‐DCNN methods, the recognition accuracy is improved by 3.5% and 4%. [ABSTRACT FROM AUTHOR]

Blind Source Separation (BSS) of complex signals composed of radar, communication and jamming signals is the first step in an integrated electronic system, which requires higher accuracy of separation. However, the traditional Single‐Channel Blind Source Separation (SCBSS) method has low separation accuracy and poor robustness. Aiming at this problem, this paper proposes a SCBSS method based on spatial‐temporal fusion deep learning model. This is a deep neural network model, which realizes spatial‐temporal of mixed signals by integrating Convolutional Neural Network (CNN) and Bidirectional Long Short‐Term Memory Network (BiLSTM). Convolutional Neural Network is used to extract spatial features from input sequences, and BiLSTM is used to mine timing rules of signals. A batch normalisation layer and a dropout layer are added to improve stability and prevent overfitting. The experiments show that the average similarity coefficient of the separated signals is above 0.99 and the Signal‐Distortion Ratio (SDR) is up to 27 dB without noise. When the Signal‐Noise Ratio is 0–20 dB and Jamming‐Signal Ratio is 15 dB, the SDR is 5–30 dB higher than the traditional methods and the single network structure deep learning methods. [ABSTRACT FROM AUTHOR]

Segmenting deep brain structures from magnetic resonance images is important for patient diagnosis, surgical planning, and research. Most current state‐of‐the‐art solutions follow a segmentation‐by‐registration approach, where subject magnetic resonance imaging (MRIs) are mapped to a template with well‐defined segmentations. However, registration‐based pipelines are time‐consuming, thus, limiting their clinical use. This paper uses deep learning to provide a one‐step, robust, and efficient deep brain segmentation solution directly in the native space. The method consists of a preprocessing step to conform all MRI images to the same orientation, followed by a convolutional neural network using the nnU‐Net framework. We use a total of 14 datasets from both research and clinical collections. Of these, seven were used for training and validation and seven were retained for testing. We trained the network to segment 30 deep brain structures, as well as a brain mask, using labels generated from a registration‐based approach. We evaluated the generalizability of the network by performing a leave‐one‐dataset‐out cross‐validation, and independent testing on unseen datasets. Furthermore, we assessed cross‐domain transportability by evaluating the results separately on different domains. We achieved an average dice score similarity of 0.89 ± 0.04 on the test datasets when compared to the registration‐based gold standard. On our test system, the computation time decreased from 43 min for a reference registration‐based pipeline to 1.3 min. Our proposed method is fast, robust, and generalizes with high reliability. It can be extended to the segmentation of other brain structures. It is publicly available on GitHub, and as a pip package for convenient usage. [ABSTRACT FROM AUTHOR]

This paper introduce an efficient object detection network named Trident‐You Only Look Once (YOLO), which is designed for mobile devices with limited computing power. The new architecture is improved based on YOLO v4‐tiny. The authors redesign the network structure and propose a trident feature pyramid network (Trident‐FPN), which can improve the precision and recall of lightweight object detection. Specifically, Trident‐FPN increases the computational complexity by only a small amount of floating point operations per second (FLOPs) and obtains a multi‐scale feature map of the model, which significantly lightweight object detection performance. To enlarge the receptive field of the network with the fewest FLOPs, this paper redesign the receptive field block (RFB) and spatial pyramid pooling (SPP) layer and propose tinier cross‐stage partial RFBs and smaller cross‐stage partial SPPs. This paper present extensive experiments, and Trident‐YOLO shows strong performance compared to that of other popular models on the PASCAL VOC and MS COCO. On the MS COCO and PASCAL VOC 2007 test sets, the mean average precision (mAP) of Trident‐YOLO improved by 4.5% and 5.0%, respectively. Trident‐YOLO also reduce the network size by more than 54.4% compared to YOLO v4‐tiny. With a 23.7% FLOP reduction, the FPS is improved by 1.9 on an Nvidia Jetson Xavier NX. [ABSTRACT FROM AUTHOR]

Structural damage detection (SDD) still suffers from environmental uncertainties or modeling errors, causing a gap between the numerical model and the real structure. It results in performance degradation in the application of many model‐based methods, which are usually designed on a numerical model and needed to be applied to a real structure. Such a situation is defined as a cross‐domain SDD problem in this work. This paper aims to address the cross‐domain SDD problem by designing a feature‐extractor to generate both damage‐sensitive and domain‐invariant features, instead of trying to reduce the gap, as the traditional methods do. A domain adaptation (DA) neural network is designed and trained on the data from both the numerical model and the real structure at the same time. In addition, no damage label of the real structure is needed. Both numerical and laboratory experiments show that the proposed method has excellent performance and outperforms the baseline model, a traditional convolutional neural network (CNN). This paper provides a new methodology to solve the cross‐domain SDD problem, that is, to learn better features instead of just trying to reduce the gap. [ABSTRACT FROM AUTHOR]

This paper plans to develop an intelligent super resolution model with the linkage of Wavelet lifting scheme and Deep learning algorithm. Before initiating the resolution procedure, the entire HR images are converted into Low Resolution (LR) images using bicubic interpolation‐based downsampling and upsampling. Further, the Wavelet lifting scheme helps to generate the four subbands of each image like LR wavelet Sub‐Bands for LR images, and High Resolution (HR) wavelet Sub‐Bands for HR images. The residual image is generated by taking the difference between the LR wavelet Sub‐Bands and HR wavelet Sub‐Bands images. The proposed model involves two main phases: Training phase and Testing. The training phase trains the residual image of all images by Deep Convolutional Neural Network with LR wavelet Sub‐Bands as input and residual image as target. On the other hand, in testing phase, the LR wavelet Sub‐Bands query image is subjected to Deep Convolutional Neural Network, which outputs the concerned residual image. This generated residual image is summed with LR wavelet Sub‐Bands image, followed by inverse wavelet lifting scheme to obtain the final super resolution image. The main contribution of this paper is to improve the conventional Deep Convolutional Neural Network by optimizing the number of hidden layer, and hidden neurons using modified Whale Optimization Algorithm called Average Fitness Enabled Whale Optimization Algorithm by considering the objective of maximizing the Peak Signal‐to‐Noise Ratio. Finally, the proposed method achieves an improved quality of the results which is comparable the existing models. [ABSTRACT FROM AUTHOR]

In the process of detecting materials inside a tray in an automated production line, it is necessary to detect not only the known materials and blank space in the designated area, but also the unknown materials misplaced inside the tray. However, the supervised detection algorithm based on deep learning can only detect the known and blank material areas. Therefore, this paper proposed a phased material detection. The first stage is to detect the tray and then identify the material area in the second stage. In order to improve the tray detection accuracy during the first stage under the condition of a high intersection ratio, an improved YOLOv5s tray detection method is proposed. The structure of YOLOv5s is improved using the SENet. This paper proposes to use the rich geometric information of the shallow network and the high‐level semantic information to integrate the bypass features. MAP@0.5:0.95 of the improved model increased from 95.7% to 96.6% and MAP@0.95 from 78.5% to 90.8%. The challenge of detecting unknown wrong materials on a tray can be resolved through the recognition of material area segmentation images processed by using the improved pre‐detection algorithm, together with the relative position reference between the material area and the tray. The experimental results showed that the improved method proposed meets the industrial detection requirements with an overall recognition accuracy of 91% within a 250 ms detection interval. [ABSTRACT FROM AUTHOR]

In the power system operational/planning studies, it is a crucial task to provide the load consumption information in the look‐ahead times. The huge variation of the power system infrastructure in recent years has led to significant changes in the consumers' consumption pattern. Therefore, short‐term load forecasting (STLF) is transformed to a more complicated problem in recent years. To address this issue, this paper proposes a graph‐based deep neural network to capture full spatial‐temporal features and be able to oversee high volatility time series including load sequence. The proposed spatial deep learning structure benefits from learning the spatial feature using Gabor filter‐oriented layers and full understanding the temporal behaviour based on bidirectional networks. The designed learning‐based system is developed as a graph‐based learning system to improve the accuracy considering the meteorological information behaviour. To verify the performance of the designed deep graph network, the actual load data of Shiraz, Iran, is used. Besides, to demonstrate the superiority and effectiveness of the proposed, the designed deep graph network is compared with three well‐known shallow and deep networks in different cases including yearly performance, seasonal performance, and sensitivity analysis on the meteorological data. [ABSTRACT FROM AUTHOR]