Search

High temporal resolution rainfall data are among the most demanded resources for water resource engineers. In modern times, this need has only multiplied day by day due to the need for training large parameter-heavy models for the prediction of climatic features, analysis of extreme rainfall, etc. However, the availability of such high temporal resolution data is low, which can cause hindrances in research or development projects in several regions. It is therefore imperative to find newer and better models for the disaggregation of rainfall data from lower to higher temporal resolutions, such as a model that uses deep learning neural networks. The main issue with such a model is the requirement for historical rainfall data at different time scales for training, testing, and validating prior to use in practical scenarios, data that may not always be available for all regions necessary. In this paper, an attempt has been to test the accuracy and applicability of pretrained models for the purpose of disaggregating rainfall in other geographical locations, thus reducing the requirement for historical rainfall data for training and validation purposes. A large data set comprising rainfall data from 68 rain gauge stations across the Indian subcontinent has been used to test models pretrained using rainfall data from seven major stations in India (Bikaner, Chennai, Cherrapunji, Delhi, Kolkata, Mumbai, and Mangalore). The pretrained models are tested in their ability to conserve extreme rainfall characteristics by comparing intensity–duration–frequency (IDF) curves generated from observed and disaggregated rainfall, further which the errors in these IDF curves are used to generate heatmaps for the country using the inverse distance weighted interpolation method. At the end of this paper, a map is provided that covers the entire country of study, detailing that a pretrained model can be used for a certain region based on its accuracy of disaggregation and proximity to the city of pretraining data. [ABSTRACT FROM AUTHOR]

The technology of big data analysis and artificial intelligence deep learning has been actively cross-combined with various fields to increase the effect of its original low single field. Precision components commonly used in electronic products use changes in the conductivity of semiconductors to process information. This study aims to review key milestones and recent developments in the semiconductor industry using artificial intelligence methods. For this systematic review, we searched academic networks between 2015 and 2022, including Nature, Elsevier, Springer, Taylor & Francis Online, Multidisciplinary Digital Publishing Institute, and the Institute of Electrical and Electronics Engineers. The literature reviewed is based on conference proceedings and journal articles, specifically covering the key achievements of the discussion paper, the key technologies used, experimental results, opportunities, and future research pathways. After searching on an academic website, we selected six major studies. In five of these studies, visual object detection, surface defect detection, machine production scheduling application, fault diagnosis and prediction, and monitoring of the manufacturing process were made using artificial neural networks, machine learning methods, and hybrid models. In addition, the studies covered independent, single methods or used more than two types of technologies for performance comparison. Finally, we reviewed the strengths and weaknesses of the literature. We also analysed various datasets, acquisition systems, and experimental scenarios. The review shows that as the number of studies conducted in manufacturing continues to increase, more research is needed to unearth key information that is often overlooked, all of which are challenges in refining science and overcoming real-world scenarios. [ABSTRACT FROM AUTHOR]

Neural network models enjoy success on language tasks related to Web documents, including news and Wikipedia articles. However, the characteristics of scientific publications pose specific challenges that have yet to be satisfactorily addressed: the discourse structure of scientific documents crucial in scholarly document processing (SDP) tasks, the interconnected nature of scientific documents, and their multimodal nature. We survey modern neural network learning methods that tackle these challenges: those that can model discourse structure and their interconnectivity and use their multimodal nature. We also highlight efforts to collect large-scale datasets and tools developed to enable effective deep learning deployment for SDP. We conclude with a discussion on upcoming trends and recommend future directions for pursuing neural natural language processing approaches for SDP. [ABSTRACT FROM AUTHOR]

Over the past few years, there has been a growing interest in this field of Artificial Intelligence (AI). This review paper aims to examine the extent to which AI has been adopted across various domains in different regions of the world. Specifically, this study seeks to explore the effectiveness of AI in areas such as banking, governance, judiciary, and agriculture. While AI's impact on these areas has been studied before, this research focuses on aspects that have not been given much needed attention so far. The study will begin by offering an overview of AI's development from a regional perspective, analyzing 63 articles from leading journals and institutions to identify the most commonly used terminologies, theories, and technologies. The increasing computational power has played a crucial role in the advancement of AI. However, the role of AI in governance, judiciary, and legislature system has not been explored in depth in previous research. Furthermore, most studies tend to focus on single domain, whereas this paper aims to examine several domains and their relationship to AI. The study presents several AI-driven techniques, including Deep Neural Net (DNN), Artificial Neural Net (ANN), Candlestick Plots, Natural Language Processing (NLP), Explainable Deep Neural Net (xDNN), Attentional Maps, Digital Inventory, Deep Learning among others, that can be applied to enhance society's performance in different domains. The research also delves deeper into the judiciary domain, proposing specific AI techniques that could be useful in this field. Additionally, tables summarizing different AI tactics relevant to each domain are presented. [ABSTRACT FROM AUTHOR]

Municipal solid waste (MSW) management has become a highly challenging issue for many countries at different levels of development since the growth of population and urbanisation has resulted in a large increase in MSW generation. It is difficult to forecast the solid waste generated and its characteristics due to the non-linear nature of the MSW system, hence the importance of introducing artificial intelligence (AI) techniques and machine learning (ML) methods. This paper discusses a systematic literature review (SLR) on the application of AI techniques in MSW management, including waste generation prediction, waste collection and transportation and waste treatment and final disposal. The study reviewed and analysed the research studies published between 2000 and 2021, and investigated the current challenges faced by researchers in implementing AI approaches in the MSW system. It was concluded that artificial neural networks are the most used approach in various MSW-related problems. However, the lack and reliability of data are limiting the advancement of AI techniques in this field. Additionally, most studies claimed that their results are accurate and can be implemented in real-life scenarios, with an absence of a clear baseline to assess the performance of the adopted approaches. The detailed gaps and future suggestions for AI techniques in MSW systems are also discussed for further research. [ABSTRACT FROM AUTHOR]

Background: The discovery of anticancer drug combinations is a crucial work of anticancer treatment. In recent years, pre-screening drug combinations with synergistic effects in a large-scale search space adopting computational methods, especially deep learning methods, is increasingly popular with researchers. Although achievements have been made to predict anticancer synergistic drug combinations based on deep learning, the application of multi-task learning in this field is relatively rare. The successful practice of multi-task learning in various fields shows that it can effectively learn multiple tasks jointly and improve the performance of all the tasks. Methods: In this paper, we propose MTLSynergy which is based on multi-task learning and deep neural networks to predict synergistic anticancer drug combinations. It simultaneously learns two crucial prediction tasks in anticancer treatment, which are synergy prediction of drug combinations and sensitivity prediction of monotherapy. And MTLSynergy integrates the classification and regression of prediction tasks into the same model. Moreover, autoencoders are employed to reduce the dimensions of input features. Results: Compared with the previous methods listed in this paper, MTLSynergy achieves the lowest mean square error of 216.47 and the highest Pearson correlation coefficient of 0.76 on the drug synergy prediction task. On the corresponding classification task, the area under the receiver operator characteristics curve and the area under the precision–recall curve are 0.90 and 0.62, respectively, which are equivalent to the comparison methods. Through the ablation study, we verify that multi-task learning and autoencoder both have a positive effect on prediction performance. In addition, the prediction results of MTLSynergy in many cases are also consistent with previous studies. Conclusion: Our study suggests that multi-task learning is significantly beneficial for both drug synergy prediction and monotherapy sensitivity prediction when combining these two tasks into one model. The ability of MTLSynergy to discover new anticancer synergistic drug combinations noteworthily outperforms other state-of-the-art methods. MTLSynergy promises to be a powerful tool to pre-screen anticancer synergistic drug combinations. [ABSTRACT FROM AUTHOR]

Handwritten digit recognition is a complex problem that has stumped even the brilliant minds of this century. Getting precise results from different handwritten samples has been a challenge that needs to be addressed due to the occurrence of this issue in several sectors like document verification, post mail, deciphering, etc. Hence, we introduce our paper as a response to the requirement of an accurate model that can acutely recognize and then predict the handwriting of a variety of individuals with ease. Our model aims to do number recognition through the implementation of neural networks. We tested out different models with each optimizer to verify which model provided the best performance and with which optimizer. Optimizers are an inherent part of Deep learning, and they are used to upgrade the weights, so the model can learn accordingly and get a more accurate system. Instead of just comparing the performance of various optimizers with only one model, we compared different model performances, while trying to select the optimizer that would best suit that learning model. Rigorous training and experimentalizing have resulted in an accuracy of 98.55% for an ANN model employed with an Adagrad optimizer. [ABSTRACT FROM AUTHOR]

Infrared small target detection (ISTD) is the main research content for defense confrontation, long-range precision strikes and battlefield intelligence reconnaissance. Targets from the aerial view have the characteristics of small size and dim signal. These characteristics affect the performance of traditional detection models. At present, the target detection model based on deep learning has made huge advances. The You Only Look Once (YOLO) series is a classic branch. In this paper, a model with better adaptation capabilities, namely ISTD-YOLOv7, is proposed for infrared small target detection. First, the anchors of YOLOv7 are updated to provide prior. Second, Gather-Excite (GE) attention is embedded in YOLOv7 to exploit feature context and spatial location information. Finally, Normalized Wasserstein Distance (NWD) replaces IoU in the loss function to alleviate the sensitivity of YOLOv7 for location deviations of small targets. Experiments on a standard dataset show that the proposed model has stronger detection performance than YOLOv3, YOLOv5s, SSD, CenterNet, FCOS, YOLOXs, DETR and the baseline model, with a mean Average Precision (mAP) of 98.43%. Moreover, ablation studies indicate the effectiveness of the improved components. [ABSTRACT FROM AUTHOR]

It is of great significance to compress neural network models so that they can be deployed on resource-constrained embedded mobile devices. However, due to the lack of theoretical guidance for non-salient network components, existing model compression methods are inefficient and labor-intensive. In this paper, we propose a new pruning method to achieve model compression. By exploring the rank ordering of the feature maps of convolutional layers, we introduce the concept of sensitive layers and treat layers with more low-rank feature maps as sensitive layers. We propose a new algorithm for finding sensitive layers while using reinforcement learning deterministic strategies to automate pruning for insensitive layers. Experimental results show that our method achieves significant improvements over the state-of-the-art in floating-point operations and parameter reduction, with lower precision loss. For example, using ResNet-110 on CIFAR-10 achieves a 62.2% reduction in floating-point operations by removing 63.9% of parameters. When testing ResNet-50 on ImageNet, our method reduces floating-point operations by 53.8% by deleting 39.9% of the parameters. [ABSTRACT FROM AUTHOR]

The existing deep learning models have problems such as large weight parameters and slow inference speed of equipment. In practical applications such as fire detection, they often cannot be deployed on equipment with limited resources due to the huge amount of parameters and low efficiency. In response to this problem, this paper proposes a lightweight smoke detection model based on the convolutional attention mechanism module. The model is based on the YOLOv5 lightweight framework. The backbone network draws on the GhostNet design idea, replaces the CSP structure of the FPN and head layers with the GhostBottleNeck module, adds a convolutional attention mechanism module to the backbone network layer, and uses the CIoU loss function to improve the regression accuracy. Using YOLOv5s as the benchmark model, the parameter amount of the proposed lightweight neural network model is 2.75 M, and the floating-point calculation amount is 2.56 G, which is much lower than the parameter amount and calculation amount of the benchmark model. Tested on the public fire dataset, compared with the traditional deep learning algorithm, the model proposed in the paper has better detection performance and the detection speed is significantly better than the benchmark model. Tested under the unquantized simulator, the speed of the proposed model to detect a single picture is 60 ms, which can meet the requirements of real-time engineering applications. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the traditional deep learning algorithms have unsatisfactory speech enhancement effect under low Signal-to-Noise Ratio (SNR) conditions, this paper improves Deep Neural Network (DNN) and Multiresolution Residual U Network (MultiResU_Net) to discuss the speech enhancement effect. In this paper, the Minimum Mean-Square Error Short-Time Spectral Amplitude Estimator (MMSE-STSA) is used to reduce the noise of the Multi-Resolution Cochleagram (MRCG) features that including the local and global characteristics of speech and using the improved features Improved Multi-Resolution Cochleagram (I-MRCG) as the input of Skip-DNN network to build Skip-DNN speech enhancement model. The short-time Fourier transformed features of the speech signal at low SNR are used as the input of MultiResU_Net network and the output features of residual path and upsampling are rearranged by the hybrid channel to build an improved MultiResU_Net speech enhancement model. Using LibriSpeech ASR corpus, under different SNR conditions, analyze the speech enhancement effects of different network models and different features as the input of the Skip-DNN model. The experimental results show that when the SNR is -5dB, the perceptual evaluation effect of the average speech quality is better that I-MRCG is used as the feature input of the Skip-DNN model. The average short-time objective intelligibility (STOI), the average perceptual evaluation of speech quality (PESQ) and Source-to-Distortion Ratio (SDR) of the improved MultiResU_Net speech enhancement model are higher than other models. It can be seen that both the improved Skip-DNN and MultiResU_Net proposed in this paper have better speech enhancement effect, especially the improved MultiResU_Net network is more suitable for speech enhancement at low SNR. [ABSTRACT FROM AUTHOR]

Introduction: The emergence of automated machine learning or AutoML has raised an interesting trend of no-code and low-code machine learning where most tasks in the machine learning pipeline can possibly be automated without support from human data scientists. While it sounds reasonable that we should leave repetitive trial-and-error tasks of designing complex network architectures and tuning a lot of hyperparameters to AutoML, leading research using AutoML is still scarce. Thereby, the overall purpose of this case study is to investigate the gap between current AutoML frameworks and practical machine learning development. Case description: First, this paper confirms the increasing trend of AutoML via an indirect indicator of the numbers of search results in Google trend, IEEE Xplore, and ACM Digital Library during 2012–2021. Then, the three most popular AutoML frameworks (i.e., Auto-Sklearn, AutoKeras, and Google Cloud AutoML) are inspected as AutoML's representatives; the inspection includes six comparative aspects. Based on the features available in the three AutoML frameworks investigated, our case study continues to observe recent machine learning research regarding the background of image-based machine learning. This is because the field of computer vision spans several levels of machine learning from basic to advanced and it has been one of the most popular fields in studying machine learning and artificial intelligence lately. Our study is specific to the context of image-based road health inspection systems as it has a long history in computer vision, allowing us to observe solution transitions from past to present. Discussion and evaluation: After confirming the rising numbers of AutoML search results in the three search engines, our study regarding the three AutoML representatives further reveals that there are many features that can be used to automate the development pipeline of image-based road health inspection systems. Nevertheless, we find that recent works in image-based road health inspection have not used any form of AutoML in their works. Digging into these recent works, there are two main problems that best conclude why most researchers do not use AutoML in their image-based road health inspection systems yet. Firstly, it is because AutoML's trial-and-error decision involves much extra computation compared to human-guided decisions. Secondly, using AutoML adds another layer of non-interpretability to a model. As these two problems are the major pain points in modern neural networks and deep learning, they may require years to resolve, delaying the mass adoption of AutoML in image-based road health inspection systems. Conclusions: In conclusion, although AutoML's utilization is not mainstream at this moment, we believe that the trend of AutoML will continue to grow. This is because there exists a demand for AutoML currently, and in the future, more demand for no-code or low-code machine learning development alternatives will grow together with the expansion of machine learning solutions. Nevertheless, this case study focuses on selected papers whose authors are researchers who can publish their works in academic conferences and journals. In the future, the study should continue to include observing novice users, non-programmer users, and machine learning practitioners in order to discover more insights from non-research perspectives. [ABSTRACT FROM AUTHOR]

Face mask-wearing detection is of great significance for safety protection during the epidemic. Aiming at the problem of low detection accuracy due to the problems of occlusion, complex illumination, and density in mask-wearing detection, this paper proposes a neural network model based on the loss function and attention mechanism for mask-wearing detection in complex environments. Based on YOLOv5s, we first introduce an attention mechanism in the feature fusion process to improve feature utilization, study the effect of different attention mechanisms (CBAM, SE, and CA) on improving deep network models, and then explore the influence of different bounding box loss functions (GIoU, CIoU, and DIoU) on mask-wearing recognition. CIoU is used as the frame regression loss function to improve the positioning accuracy. By collecting 7,958 mask-wearing images and a large number of images of people without masks as a dataset and using YOLOv5s as the benchmark model, the mAP of the model proposed in the paper reached 90.96% on the validation set, which is significantly better than the traditional deep learning method. Mask-wearing detection is carried out in a real environment, and the experimental results of the proposed method can meet the daily detection requirements. [ABSTRACT FROM AUTHOR]

The piano, as the king of playing instruments, is the most popular instrument for people to learn to play. Learning to play the piano, on the other hand, necessitates professional instruction and a lot of practice. People do not have enough time for systematic training because of the fast pace of life. At the same time, a lack of professional piano teachers and high tuition fees discourage piano students. If the computer can recognize and evaluate the learner's piano music in real time, the learner will be able to identify and correct errors in real time. There are currently some music recognition technologies, but the majority of them have the following flaws: first and foremost, the recognition accuracy is poor. Second, the identification process is slow and not real-time. Based on the existing problems, this paper proposes a mobile-based music recognition method. The main work of this paper is as follows: (1) a deep neural network (DNN) is applied to the recognition of piano playing music. The use of deep learning models improves the accuracy of music recognition. (2) In order to make the identification of music easier to use, a mobile application is developed in this paper. The app can be installed on mobile phones and tablets. It can input songs in real-time or offline, outputting misplayed notes and scoring the entire composition. In order to evaluate the effect of this study on music recognition, the experimental part uses multiple models for comparison. The experimental results show that the research in this paper is feasible and effective. [ABSTRACT FROM AUTHOR]

To ensure the safety and rational use of bridge traffic lines, the existing bridge structural damage detection models are not perfect for feature extraction and have difficulty meeting the practicability of detection equipment. Based on the YOLO (You Only Look Once) algorithm, this paper proposes a lightweight target detection algorithm with enhanced feature extraction of bridge structural damage. The BIFPN (Bidirectional Feature Pyramid Network) network structure is used for multi-scale feature fusion, which enhances the ability to extract damage features of bridge structures, and uses EFL (Equalized Focal Loss) to optimize the sample imbalance processing mechanism, which improves the accuracy of bridge structure damage target detection. The evaluation test of the model has been carried out in the constructed BDD (Bridge Damage Dataset) dataset. Compared with the YOLOv3-tiny, YOLOv5S, and B-YOLOv5S models, the mAP@.5 of the BE-YOLOv5S model increased by 45.1%, 2%, and 1.6% respectively. The analysis and comparison of the experimental results prove that the BE-YOLOv5S network model proposed in this paper has a better performance and a more reliable performance in the detection of bridge structural damage. It can meet the needs of bridge structure damage detection engineering with high requirements for real-time and flexibility. [ABSTRACT FROM AUTHOR]

With the continuous deepening of enterprise system reform and the rapid development of national economy, enterprises are facing the great challenge of market competition. In the new market and social environment, the role of human resource management in enterprises becomes particularly important. To further improve the level of enterprise human resources strategic management has become an urgent problem to be solved. In the process of human resource management, enterprises are faced with complex and changeable environment and other influencing factors. Therefore, in the human resource information retrieval, this paper uses the method of deep learning to screen human resource management indicators and constructs the human resource management index system of power supply enterprises. In this paper, the nonlinear characteristics of neural network are used to establish a deep neural network human resource cross-media fusion model, which provides an operational method for enterprise human resource management. The human resource allocation relationship of enterprises is predicted, and the influencing factors and trends of personnel post-matching are analyzed. The demand forecasting results show that the neural network depth has a good fit with the enterprise staff, and the actual forecasting error is less than 3.0. It can accurately predict the human resource allocation of enterprises, improve the scientificity and effectiveness of human resource strategic decision-making, and make enterprises better adapt to the requirements of market economy. This will be of practical significance to the modernization of enterprise management. [ABSTRACT FROM AUTHOR]

A comprehensive systematic study was carried out in order to identify various deep learning methods developed and used for predicting student academic performance. Predicting academic performance allows for the implementation of various preventive and supportive measures earlier in order to improve academic performance and reduce failure and dropout rates. Although machine learning schemes were once popular, deep learning algorithms are now being investigated to solve difficult predictions of student performance in larger datasets with more data attributes. Deep neural network prediction methods with clear modelling and parameter measurements formulated on publicly available and recognised datasets are the focus of the research. Widely used for academic performance prediction, backpropagation algorithms have been trained and tested with various datasets, especially those related to learning management systems (LMS) and massive open online courses (MOOC). The most widely used prediction method appears to be the standard artificial neural network approach. The long-short-term memory (LSTM) approach has been reported to achieve an accuracy of around 87 percent for temporal student performance data. The number of papers that study and improve this method shows that there is a clear rise in deep learning-based academic performance prediction over the last few years. [ABSTRACT FROM AUTHOR]

The accurate prediction of sea surface temperature (SST) is the basis for our understanding of local and global climate characteristics. At present, the existing sea temperature prediction methods fail to take full advantage of the potential spatial dependence between variables. Among them, graph neural networks (GNNs) modeled on the relationships between variables can better deal with space–time dependency issues. However, most of the current graph neural networks are applied to data that already have a good graph structure, while in SST data, the dependency relationship between spatial points needs to be excavated rather than existing as prior knowledge. In order to predict SST more accurately and break through the bottleneck of existing SST prediction methods, we urgently need to develop an adaptive SST prediction method that is independent of predefined graph structures and can take full advantage of the real temporal and spatial correlations hidden indata sets. Therefore, this paper presents a graph neural network model designed specifically for space–time sequence prediction that can automatically learn the relationships between variables and model them. The model automatically extracts the dependencies between sea temperature multi-variates by embedding the nodes of the adaptive graph learning module, so that the fine-grained spatial correlations hidden in the sequence data can be accurately captured. Figure learning modules, graph convolution modules, and time convolution modules are integrated into a unified end-to-end framework for learning. Experiments were carried out on the Bohai Sea surface temperature data set and the South China Sea surface temperature data set, and the results show that the model presented in this paper is significantly better than other sea temperature model predictions in two remote-sensing sea temperature data sets and the surface temperature of the South China Sea is easier to predict than the surface temperature of the Bohai Sea. [ABSTRACT FROM AUTHOR]

Object tracking based on Siamese networks has achieved great success in recent years, but increasingly advanced trackers are also becoming cumbersome, which will severely limit deployment on resource-constrained devices. To solve the above problems, we designed a network with the same or higher tracking performance as other lightweight models based on the SiamFC lightweight tracking model. At the same time, for the problems that the SiamFC tracking network is poor in processing similar semantic information, deformation, illumination change, and scale change, we propose a global attention module and different scale training and testing strategies to solve them. To verify the effectiveness of the proposed algorithm, this paper has done comparative experiments on the ILSVRC, OTB100, VOT2018 datasets. The experimental results show that the method proposed in this paper can significantly improve the performance of the benchmark algorithm. [ABSTRACT FROM AUTHOR]

In this paper, a literature overview of propagation models based on the application of deep learning is presented. Given the shortcomings of traditional empirical and deterministic models, papers dealing with the formation of propagation models are increasingly turning to modern solutions such as deep learning. This paper discusses the classification of propagation models based on deep learning techniques based on their ability to make predictions independently or in combination with a traditional model. Different architectures of deep learning models that are most commonly used in the creation of propagation models are analyzed, such as deep feedforward neural networks, convolutional neural networks and generative adversarial networks. The basic differences arising from different deep learning models and types of input data are analyzed, as well as their impact on the need for expert knowledge in the selection of data that appear as elements in the vector representing input. [ABSTRACT FROM AUTHOR]

With the current exchange and communication between different countries becoming more and more frequent, the language conversion of different countries has become a difficult problem. The analysis of a series of problems in cross-language discourse conversion, the study of the discourse conversion path, and innovation motivation based on the deep learning theory of cross-language transfer, it has theoretical and practical significance. This paper aims at the technical difficulties in speech conversion methods to effectively utilize the local mode information of signal time spectrum and the long-term correlation of speech signal. A discourse conversion method based on convolutional recurrent neural network model is proposed. In the model, the extended convolutional neural network is used to model the long-term correlation of speech signals. In the part of speech fundamental frequency estimation, the prosodic information generated by the decomposition of the fundamental frequency by continuous wavelet transform is used as the training target of the fundamental frequency estimation model. The experimental results show that the speech transformation method based on the convolutional cyclic network model proposed in this paper has better quality and intelligibility than the speech transformed by the contrast method. [ABSTRACT FROM AUTHOR]

Purpose: Current research on the aging of bony orbit is usually done manually, which is inefficient and has a large error. In this paper, automatic segmentation of bony orbit based on deep learning and automatic calculation of the parameters of the segmented orbital contour (area and height of bony orbit) are presented. Methods: The craniofacial CT scanning data of 595 Chinese were used to carry out three-dimensional reconstruction and output the craniofacial images. The orbital contour images are obtained automatically by UNet++ segmentation network, and then the bony orbital area and height were calculated automatically by connected component analysis. Results: The automatic segmentation method has an Intersection of Union of 95.41% in craniofacial CT images. During the aging, the bony orbital area of males increased with age, while that of females decreased, and the area in male was larger than that in female (P < 0.05). The distance from equal points 10 and 40–90 to the supraorbital rim was significantly larger (P < 0.05). Except for the equal point 90, the distance from equal points to the inferior orbital rim was obviously larger (P < 0.05). In the females, the distance from equal points 50–70 to inferior orbital rim was significantly lower (P < 0.05). Conclusion: The method proposed here can automatically and accurately study image dataset of large-scale bony orbital CT imaging. UNet++ can achieve high-precision segmentation of bony orbital contours. The bony orbital area of Chinese changes with aging, and the bony orbital height changes different between males and females, which may be caused by the different position and degree of orbital bone resorption of males and females in the process of aging. [ABSTRACT FROM AUTHOR]

The June 2023 issue of the CIT Journal of Computing and Information Technology features four papers on various topics. The first paper discusses an automatic diagnosis method for short circuit faults in power measurement instruments using deep learning networks. The second paper focuses on a network intrusion detection model that uses a genetic algorithm to improve accuracy and minimize false alarms. The third paper explores a distributed framework for point cloud processing of ancient buildings, aiming to increase data compression accuracy and storage efficiency. The final paper presents a deep neural network model for detecting small targets, specifically safety helmets, in complex environments. Overall, these papers offer valuable insights and improvements in their respective fields. [Extracted from the article]

In recent years, some researchers have combined deep learning methods such as semantic segmentation with a visual SLAM to improve the performance of classical visual SLAM. However, the above method introduces the uncertainty of the neural network model. To solve the above problems, an improved feature selection method based on information entropy and feature semantic uncertainty is proposed in this paper. The former is used to obtain fewer and higher quality feature points, while the latter is used to correct the uncertainty of the network in feature selection. At the same time, in the initial stage of feature point selection, this paper first filters and eliminates the absolute dynamic object feature points in the a priori information provided by the feature point semantic label. Secondly, the potential static objects can be detected combined with the principle of epipolar geometric constraints. Finally, the semantic uncertainty of features is corrected according to the semantic context. Experiments on the KITTI odometer data set show that compared with SIVO, the translation error is reduced by 12.63% and the rotation error is reduced by 22.09%, indicating that our method has better tracking performance than the baseline method. [ABSTRACT FROM AUTHOR]

The advancement of machine learning offers a promising solution for diagnosing both the extent of leakage and the precise location of natural gas transmission pipeline leaks. While pipeline leaks often manifest through sound-related signals, leveraging machine learning for signal classification is still in its nascent stages. We propose a hybrid technological approach for detecting and pinpointing gas pipeline leaks, which is rooted in machine learning principles. In this paper, we simulate and classify the degree of pipe-line leakage by adjusting the bolt aperture. We analyze and select original data acquired through acquisition and processing using eigenvalue methods. Subsequently, we employ a deep learning algorithm for training and testing the data post feature extraction, enabling the realization of pipeline leakage diagnosis and classification. To fully leverage the information embedded in the original data, we adopt the radial basis function neural network (RBF-NN) machine learning technique. Moreover, we utilize the genetic algorithm (GA) for optimization, resulting in the generation of a radial basis function neural network model (GA-RBF) optimized through genetic algorithmic techniques. We compare the predictive performance of the GA-RBF model with that of BP-NN and RBF-NN. The diagnostic F1 scores of the GA-RBF model under various operational conditions of gas pipelines are as follows: 98.30%, 97.50%, 95.10%, and 95.80%, respectively, with an overall accuracy of 96.68%. The diagnostic outcomes align well with theoretical expectations, demonstrating the superior performance of the GA-RBF model in gas pipe-line leakage diagnosis. [ABSTRACT FROM AUTHOR]

Malware remains a persistent threat to industrial operations, causing disruptions and financial losses. Traditional malware detection approaches struggle with the increasing complexity of false positives and negatives. However, existing Intrusion Detection Systems (IDSs) often lack the capability to assess the severity of detected malware, crucial for effective threat mitigation. This paper presents a novel model, MAlware DEtection and Severity Analysis for eNcrypted Traffic (MADESANT), designed to detect and analyze malware severity in encrypted traffic data. MADESANT combines Deep Learning (DL)-based intrusion detection with Machine Learning (ML)-based severity analysis, specifically customized for the minutiae of IoT systems and assets. Notably, MADESANT introduces a cascading model integrating a Cascading Forward Back Propagation Neural Network (CFBPNN) with the J48 tree to systematically assess risk factors in network traffic. Our assessment, conducted on diverse encrypted datasets including UNSW-NB15, IoT23, and XIIoTID, highlights the remarkable efficacy of MADESANT. Impressively, it achieves a flawless 0% false positive rate in detecting binary attack instances, surpassing benchmarks set by conventional models. Additionally, MADESANT excels in accurately estimate malware severity, providing invaluable insights into the factors contributing to the risk. To further validate its efficiency, we compared MADESANT against prevalent Neural Network models like FeedForward and Recurrent Neural Networks, with MADESANT emerging as the superior choice. The experimentation encompasses both the entire dataset and subsets generated through meticulous risk factor analysis. These results underscore MADESANT's prowess in not only identifying malware but also in evaluating its potential impact, signifying a significant leap forward in industrial cybersecurity. [ABSTRACT FROM AUTHOR]

The security issues with mobile devices have received more attention as a result of the development of mobile Internet technology and the adoption of mobile intelligent terminal devices. It is becoming more crucial to quickly and effectively identify and remove harmful applications from systems in order to protect user data and personal devices. The Dalvik bytecode, permission applications, and system calls of Android apps are the main targets of the current Android malware analysis approaches. However, in recent years, an increasing amount of Android malware conceals harmful code in native code. The method for using program gene technology to identify malware on the Android platform is presented in this research. This method extracts executable library files from the Native layer binary executable files of Android programs and disassembles the library files to obtain program genes. Then, the programs' genes perform feature screening by the information gain method and next use Word2Vec to express the semantic abstraction of the screened features. Finally, the screened features were used in deep neural network models for training and detection. The experimental results demonstrate that compared with KNN, SVM, and other machine learning algorithms, the deep neural network model is more effective and the detection accuracy reaches up to 97.51%. Thus, it confirmed the feasibility of the Android malicious program detection method based on program genes in this paper. [ABSTRACT FROM AUTHOR]

Deep learning (DL) and convolutional neural networks (CNNs) have achieved state-of-the-art performance in many medical image analysis tasks. Histopathological images contain valuable information that can be used to diagnose diseases and create treatment plans. Therefore, the application of DL for the classification of histological images is a rapidly expanding field of research. The popularity of CNNs has led to a rapid growth in the number of works related to CNNs in histopathology. This paper aims to provide a clear overview for better navigation. In this paper, recent DL-based classification studies in histopathology using strongly annotated data have been reviewed. All the works have been categorized from two points of view. First, the studies have been categorized into three groups according to the training approach and model construction: 1. fine-tuning of pre-trained networks for one-stage classification, 2. training networks from scratch for one-stage classification, and 3. multi-stage classification. Second, the papers summarized in this study cover a wide range of applications (e.g., breast, lung, colon, brain, kidney). To help navigate through the studies, the classification of reviewed works into tissue classification, tissue grading, and biomarker identification was used. [ABSTRACT FROM AUTHOR]

The diagnosis of different pathologies and stages of cancer using whole histopathology slide images (WSI) is the gold standard for determining the degree of tissue metastasis. The use of deep learning systems in the field of medical images, especially histopathology images, is becoming increasingly important. The training and optimization of deep neural network models involve fine-tuning parameters and hyperparameters such as learning rate, batch size (BS), and boost to improve the performance of the model in task-specific applications. Tuning hyperparameters is a major challenge in designing deep neural network models, having a large impact on the performance. This paper analyzes how the parameters and hyperparameters of a deep learning architecture affect the classification of colorectal cancer (CRC) histopathology images using the well-known VGG19 model. This paper also discusses the pre-processing of these images, such as the use of color normalization and stretching transformations on the data set. Among these hyperparameters, the most important neural network hyperparameter is the learning rate (LR). In this paper, different strategies for the optimization of LR are analyzed (both static and dynamic) and a new experiment based on the variation of LR is proposed (the relevance of dynamic strategies over fixed LR is highlighted), after each layer of the neural network together with decreasing variations according to the epochs. The results obtained are very remarkable, obtaining in the simulation an accurate system that achieves 96.4% accuracy on test images (for nine different tissue classes) using the triangular-cyclic learning rate. [ABSTRACT FROM AUTHOR]

Availability is one of the primary goals of smart networks, especially, if the network is under heavy video streaming traffic. In this paper, we propose a deep learning based methodology to enhance availability of video streaming systems by developing a prediction model for video streaming quality, required power consumption, and required bandwidth based on video codec parameters. The H.264/AVC codec, which is one of the most popular codecs used in video steaming and conferencing communications, is chosen as a case study in this paper. We model the predicted consumed power, the predicted perceived video quality, and the predicted required bandwidth for the video codec based on video resolution and quantization parameters. We train, validate, and test the developed models through extensive experiments using several video contents. Results show that an accurate model can be built for the needed purpose and the video streaming quality, required power consumption, and required bandwidth can be predicted accurately which can be utilized to enhance network availability in a cooperative environment. [ABSTRACT FROM AUTHOR]

Renewable energy resources can be deployed locally and efficiently using the concept of microgrids. Due to the natural uncertainty of the output power of renewable energy resources, the planning for a proper operation of microgrids can be a challenging task. In addition, the information about the loads and the power consumption of them can create benefits to increase the efficiency of the microgrids. However, electrical loads can have uncertainty due to reasons such as unpredictable behavior of the consumers. To exploit a microgrid, energy management is required at the upper level of operation and control in order to reduce the costs. One of the most important tasks of the energy management system is to satisfy the loads and, in other words, develop a plan to maintain equilibrium between the power generation and power consumption. To obtain information about the output power of renewable energy resources and power consumption, deep learning can be implemented as a powerful tool, which is able to predict the desired values. In addition, weather conditions can affect the output power of renewable energy-based resources and the behavior of the consumers and, as a result, the power consumption. So, deep learning can be deployed for the anticipation of the weather conditions. This paper will study the recent works related to deep learning, which has been implemented for the prediction of the output power of renewable energy resources (i.e., PVs and wind turbines), electrical loads, and weather conditions (i.e., solar irradiance and wind speed). In addition, for possible future directions some strategies are suggested, the most important of which is the implementation of quantum computing in cyber–physical microgrids. [ABSTRACT FROM AUTHOR]

In recent years, artificial neural networks have been widely used in the fault diagnosis of rolling bearings. To realize real-time diagnosis with high accuracy of the fault of a rolling bearing, in this paper, a bearing fault diagnosis model was designed based on the combination of VMD and ANN, which ensures a higher fault prediction accuracy with less computational time. This paper works from two aspects, including fault feature extraction and neural network structural parameter optimization to obtain an ANN bearing fault diagnosis model with high performance. The raw vibration signals of 10 fault types were divided into training, verification and testing datasets by the random step increment slip method. The variational mode decomposition method was used to decompose the raw vibration signal into several intrinsic mode functions. A new definition of the energy of each intrinsic mode function based on discrete Fourier transform and information entropy method were used as the input for the artificial neural network. Furthermore, the structural parameters of the artificial neural network were designed to obtain a high-performance neural network. The artificial neural network used in this paper had three hidden layers and 13 neurons in each hidden layer. Compared with several machine and deep learning algorithms, the artificial neural network can better fulfill the classification task of rolling bearing fault types with a mean prediction accuracy of 99.3% and computation time of 2.4 s based on a small training dataset. [ABSTRACT FROM AUTHOR]

Artificial Neural Networks (ANNs) are machine learning algorithms inspired by the structure and function of the human brain. Their popularity has increased in recent years due to their ability to learn and improve through experience, making them suitable for a wide range of applications. ANNs are often used as part of deep learning, which enables them to learn, transfer knowledge, make predictions, and take action. This paper aims to provide a comprehensive understanding of ANNs and explore potential directions for future research. To achieve this, the paper analyzes 10,661 articles and 35,973 keywords from various journals using a text-mining approach. The results of the analysis show that there is a high level of interest in topics related to machine learning, deep learning, and ANNs and that research in this field is increasingly focusing on areas such as optimization techniques, feature extraction and selection, and clustering. The study presented in this paper is motivated by the need for a framework to guide the continued study and development of ANNs. By providing insights into the current state of research on ANNs, this paper aims to promote a deeper understanding of ANNs and to facilitate the development of new techniques and applications for ANNs in the future. [ABSTRACT FROM AUTHOR]

Machine learning (ML) can enhance a dermatologist's work, from diagnosis to customized care. The development of ML algorithms in dermatology has been supported lately regarding links to digital data processing (e.g., electronic medical records, Image Archives, omics), quicker computing and cheaper data storage. This article describes the fundamentals of ML-based implementations, as well as future limits and concerns for the production of skin cancer detection and classification systems. We also explored five fields of dermatology using deep learning applications: (1) the classification of diseases by clinical photos, (2) der moto pathology visual classification of cancer, and (3) the measurement of skin diseases by smartphone applications and personal tracking systems. This analysis aims to provide dermatologists with a guide that helps demystify the basics of ML and its different applications to identify their possible challenges correctly. This paper surveyed studies on skin cancer detection using deep learning to assess the features and advantages of other techniques. Moreover, this paper also defined the basic requirements for creating a skin cancer detection application, which revolves around two main issues: the full segmentation image and the tracking of the lesion on the skin using deep learning. Most of the techniques found in this survey address these two problems. Some of the methods also categorize the type of cancer too. [ABSTRACT FROM AUTHOR]

Spotting rumors from social media and intervening early has always been a daunting challenge. In recent years, Deep neural networks have begun to discover rumors by exploring the way of rumor propagation. The existing static graph models either only focus on the spatial structure information of rumor propagation or on time series propagation information but do not effectively combine them. This paper proposes the Static Spatiotemporal Model (SSM), which first extracts the textual semantic information and constructs undirected and directed propagation trees. Then obtains spatial structure information of rumor propagation through Graph Convolutional Network and extracts time series propagation information through the Recurrent Neural Network. The extracted spatiotemporal information is enhanced using different source node information hopping. Finally, SSM uses a weighted connection ensemble to rumor classification. Experimentally validated on datasets such as Weibo and Twitter, the results show that the proposed method outperforms several state-of-the-art static graph models. To better apply SSM in early detection and characterize early concepts, this paper presents a new data collection index for early detection, which can detect events that spread faster and have more significant influence in a targeted manner. The experimental results on the new indicators further verify the superiority of SSM as it can extract sufficient information in early detection or events with fewer participants. [ABSTRACT FROM AUTHOR]

Depression is one of the most common psychological problems faced by human society. Because of less social experience, low psychological endurance, and the multiple responsibilities of future families and society, college students have become one of the most vulnerable groups to suffer from depression. This paper explores an automatic identification method to identify patients with early depression tendency through deep mining of online information of campus social platform users. First, we comprehensively analyze the common characteristics of emotion and behavior on the campus social platform for depression. Secondly, the experimental corpus is formed by preprocessing operations such as deprivation, word segmentation, and denoising of the original data. Finally, the depression recognition is transformed into a text classification problem, and a shallow support vector machine and a deep convolutional neural network model are, respectively, constructed based on the experimental corpus. Combined with the features of depression blog, the algorithm was further improved, and a dual-input convolutional neural network algorithm compatible with multiple features was proposed. The experiments showed that the recognition rate was effectively improved. [ABSTRACT FROM AUTHOR]

Folk music style classification is of great significance. Traditional folk music style classification has difficulties in feature selection, and the existing folk music style methods based on deep learning also have shortcomings. In this paper, we use artificial neural networks to classify folk music styles and transform audio signals into a sound spectrum. In this paper, we use artificial neural networks to classify folk music styles and transform audio signals into a sound spectrum to avoid the problem of manually selecting features. Further, we combine the characteristics of the music signal and a variety of music data enhancement methods to enhance the music data. The proposed model can extract elements of the sound spectrum that are more closely associated with a certain music style category. Experimental results reveal that the proposed method achieves a high accuracy rate, which verifies the effectiveness of our model. [ABSTRACT FROM AUTHOR]

The supervision of security risk level of carbofuran pesticide residues can guarantee the food quality and security of residents effectively. In order to predict the potential key risk vegetables and regions, this paper constructs a security risk assessment model, combined with the k-means++ algorithm, to establish the risk security level. Then the evaluation index value of the security risk model is predicted to determine the security risk level based on the deep learning model. The model consists of a convolutional neural network (CNN) and a long short-term memory network (LSTM) optimized by an arithmetic optimization algorithm (AOA), namely, CNN-AOA-LSTM. In this paper, a comparative experiment is conducted on a small sample data set of independently constructed security risk assessment indicators. Experimental results show that the accuracy of the CNN-AOA-LSTM prediction model based on attention mechanism is 6.12% to 18.99% higher than several commonly used deep neural network models (gated recurrent unit, LSTM, and recurrent neural networks). The prediction model proposed in this paper provides scientific reference to establish the priority order of supervision, and provides forward-looking supervision for the government. [ABSTRACT FROM AUTHOR]

Software defect prediction studies aim to predict defect-prone components before the testing stage of the software development process. The main benefit of these prediction models is that more testing resources can be allocated to fault-prone modules effectively. While a few software defect prediction models have been developed for mobile applications, a systematic overview of these studies is still missing. Therefore, we carried out a Systematic Literature Review (SLR) study to evaluate how machine learning has been applied to predict faults in mobile applications. This study defined nine research questions, and 47 relevant studies were selected from scientific databases to respond to these research questions. Results show that most studies focused on Android applications (i.e., 48%), supervised machine learning has been applied in most studies (i.e., 92%), and object-oriented metrics were mainly preferred. The top five most preferred machine learning algorithms are Naïve Bayes, Support Vector Machines, Logistic Regression, Artificial Neural Networks, and Decision Trees. Researchers mostly preferred Object-Oriented metrics. Only a few studies applied deep learning algorithms including Long Short-Term Memory (LSTM), Deep Belief Networks (DBN), and Deep Neural Networks (DNN). This is the first study that systematically reviews software defect prediction research focused on mobile applications. It will pave the way for further research in mobile software fault prediction and help both researchers and practitioners in this field. [ABSTRACT FROM AUTHOR]

Organs-at-risk contouring is time consuming and labour intensive. Automation by deep learning algorithms would decrease the workload of radiotherapists and technicians considerably. However, the variety of metrics used for the evaluation of deep learning algorithms make the results of many papers difficult to interpret and compare. In this paper, a qualitative evaluation is done on five established metrics to assess whether their values correlate with clinical usability. A total of 377 CT volumes with heart delineations were randomly selected for training and evaluation. A deep learning algorithm was used to predict the contours of the heart. A total of 101 CT slices from the validation set with the predicted contours were shown to three experienced radiologists. They examined each slice independently whether they would accept or adjust the prediction and if there were (small) mistakes. For each slice, the scores of this qualitative evaluation were then compared with the Sørensen-Dice coefficient (DC), the Hausdorff distance (HD), pixel-wise accuracy, sensitivity and precision. The statistical analysis of the qualitative evaluation and metrics showed a significant correlation. Of the slices with a DC over 0.96 (N = 20) or a 95% HD under 5 voxels (N = 25), no slices were rejected by the readers. Contours with lower DC or higher HD were seen in both rejected and accepted contours. Qualitative evaluation shows that it is difficult to use common quantification metrics as indicator for use in clinic. We might need to change the reporting of quantitative metrics to better reflect clinical acceptance. [ABSTRACT FROM AUTHOR]

The shape of a bluff body section is of high importance to its aerostatic performance. Obtaining the aerostatic performance of a specific shape based on wind tunnel tests and CFD simulations takes a lot of time, which affects evaluation efficiency. This paper proposes a novel fully convolutional neural network model that enables rapid prediction from shape to aerostatic performance. Its main innovations are: (1) The proposal of a new shape description method in which the shape is described by the combination of the wall distance field and the space coordinate field, which can efficiently express the influencing factors of the shape on the aerostatic performance. (2) A step-by-step strategy in which the pressure field is used as the model output and then the calculation of the aerostatic coefficient is proposed. Compared with the simple direct prediction of the aerostatic coefficient, the logical connection between input and output can be enhanced and the prediction accuracy can be improved. It is found that the model proposed in this paper has good prediction accuracy, and its average relative error is 9.42% compared with the CFD calculation results. Compared with the direct use of the shape as the model input, the accuracy is improved by 13.25%; compared with the direct use of the drag coefficient as the model output, the accuracy is improved by 10%. Compared with traditional CFD calculations and wind tunnel experiments, this method can be used as a fast auxiliary screening method for the optimization of the aerodynamic shapes of bluff body sections. [ABSTRACT FROM AUTHOR]

At present, in the age of computers and automation of services, deep learning (DL) technology, mainly the subset of machine learning (ML) and artificial intelligence (AI), is expressively used in innumerable domains of computer vision such as data analysis, image recognition, classification, natural language processing, and many more. It has become the foremost choice of researchers as of its effectiveness in producing decent results. This paper presents detailed and analytical literature starting from the very elementary level to the recent trends of this trending technology while focusing on the most used DL model, that is, convolutional neural network and its pretrained models for image classification and object detection. It also reviews diverse existing current literature based on this. Further, a brief introduction of AI, ML, and DL has also been presented, making the foundation for the readers. As pretrained models continuously give an upper edge to DL over ML and other technologies, 23 most popular pretrained models with their architectural diagrams have also been presented. This paper aims to summarize and analyze all the concepts used to formulate DL and its models. Also, we have emphasized more on the GoogleNet models and the entire Inception modules in detail. Finally, the fascinating applications and discussion on integral components of DL have been presented. This paper will definitely draw the attention of the students and researchers working in the area of DL and its models. [ABSTRACT FROM AUTHOR]

The rapid development of traffic industry has caused traffic congestion and environmental pollution and other problems. Traffic flow prediction and signal timing optimization under different road environments can improve the current traffic environment. In this paper, based on historical traffic data and BP neural network, according to the characteristics of different deep learning modes, SVM neural network is integrated to achieve accurate prediction of traffic flow, fully analyze the characteristics of traffic flow in different traffic environments, and accurately estimate the current traffic capacity. Meanwhile, on the basis of traffic flow prediction data and fuzzy control method, the signal timing optimization of three-phase intersections is realized to ensure the maximum capacity and the lowest average delay time of intersections. Simulation experiments show that the accuracy of the traffic flow prediction method designed in this paper is about 19.5% higher than that of the traditional mean prediction method, and fusion neural network model is more accurate than single neural network model. The optimized signal timing method can accurately control the traffic process with different phases and has obvious improvement in traffic capacity and average delay time compared with the traditional method. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNN) have fundamentally revolutionized the artificial intelligence (AI) field. The transformer model is a type of DNN that was originally used for the natural language processing tasks and has since gained more and more attention for processing various kinds of sequential data, including biological sequences and structured electronic health records. Along with this development, transformer-based models such as BioBERT, MedBERT, and MassGenie have been trained and deployed by researchers to answer various scientific questions originating in the biomedical domain. In this paper, we review the development and application of transformer models for analyzing various biomedical-related datasets such as biomedical textual data, protein sequences, medical structured-longitudinal data, and biomedical images as well as graphs. Also, we look at explainable AI strategies that help to comprehend the predictions of transformer-based models. Finally, we discuss the limitations and challenges of current models, and point out emerging novel research directions. [ABSTRACT FROM AUTHOR]

Purpose: As agricultural technology continues to develop, the scale of planting and production of date fruit is increasing, which brings higher yields. However, the increasing yields also put a lot of pressure on the classification step afterward. Image recognition based on deep learning algorithms can help to identify and classify the date fruit species, even in natural light. Method: In this paper, a deep fusion model based on whale optimization and an artificial neural network for Arabian date classification is proposed. The dataset used in this study includes five classes of date fruit images (Barhi, Khalas, Meneifi, Naboot Saif, Sullaj). The process of designing each model can be divided into three phases. The first phase is feature extraction. The second phase is feature selection. The third phase is the training and testing phase. Finally, the best-performing model was selected and compared with the currently established models (Alexnet, Squeezenet, Googlenet, Resnet50). Results: The experimental results show that, after trying different combinations of optimization algorithms and classifiers, the highest test accuracy achieved by DeepDate was 95.9%. It takes less time to achieve a balance between classification accuracy and time consumption. In addition, the performance of DeepDate is better than that of many deep transfer learning models such as Alexnet, Squeezenet, Googlenet, VGG-19, NasNet, and Inception-V3. Conclusion: The proposed DeepDate improves the accuracy and efficiency of classifying date fruits and achieves better results in classification metrics such as accuracy and F1. DeepDate provides a promising classification solution for date fruit classification with higher accuracy. To further advance the industry, it is recommended that stakeholders invest in technology transfer programs to bring advanced image recognition and AI tools to smaller producers, enhancing sustainability and productivity across the sector. Collaborations between agricultural technologists and growers could also foster more tailored solutions that address specific regional challenges in date fruit production. [ABSTRACT FROM AUTHOR]

The article discusses the use of convolutional neural networks (CNNs) in remote sensing object detection, highlighting the limitations of traditional CNN models and proposing a novel approach. Experimental results demonstrate the superiority of the proposed CNN model in detecting objects in remote sensing images. The document explains the function and structure of embedding layers, the association block, and the dynamic association module in the model. It also discusses the loss function used. The text discusses the feature signal processing of remote sensing object detection using CNNs and compares the proposed approach with other popular models. The document is a compilation of various research papers and articles related to object detection in optical remote sensing images, covering a wide range of topics and providing valuable insights and methodologies for researchers in these areas. [Extracted from the article]

In current society, escalating stress in everyday life and work creates significant concern about the potential for both physical and psychological long-term effects if unattended. To address this challenge, in this paper, we introduce a new approach designed to alleviate stress and promote well-being. Our proposed system synchronizes pointer swings with music rhythms to achieve a combination of rhythmic music engagement and interactive physical activity. This solution centers on the intersection of technology and well-being by providing users with an immersive and relaxing audio-visual experience. To achieve higher rhythm extraction precision in different genres of music, we employ wavelet transform and deep learning (DL) to adjust parameters dynamically. The controllable rotating sphere is achieved using a millimeter-wave (mmWave) radar and a thermographic camera to detect preset gestures. This system provides the interaction of rhythm and control over the rotating sphere, which can provide users with an immersive experience and contribute to a stress-relieving experience. [ABSTRACT FROM AUTHOR]

Mental stress is a common problem that affects individuals all over the world. Stress reduces human functionality during routine work and may lead to severe health defects. Early detection of stress is important for preventing diseases and other negative health-related consequences of stress. Several neuroimaging techniques have been utilized to assess mental stress, however, due to its ease of use, robustness, and non-invasiveness, electroencephalography (EEG) is commonly used. This paper aims to fill a knowledge gap by reviewing the different EEG-related deep learning algorithms with a focus on Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs) for the evaluation of mental stress. The review focuses on data representation, individual deep neural network model architectures, hybrid models, and results amongst others. The contributions of the paper address important issues such as data representation and model architectures. Out of all reviewed papers, 67% used CNN, 9% LSTM, and 24% hybrid models. Based on the reviewed literature, we found that dataset size and different representations contributed to the performance of the proposed networks. Raw EEG data produced classification accuracy around 62% while using spectral and topographical representation produced up to 88%. Nevertheless, the roles of generalizability across different deep learning models and individual differences remain key areas of inquiry. The review encourages the exploration of innovative avenues, such as EEG data image representations concurrently with graph convolutional neural networks (GCN), to mitigate the impact of inter-subject variability. This novel approach not only allows us to harmonize structural nuances within the data but also facilitates the integration of temporal dynamics, thereby enabling a more comprehensive assessment of mental stress levels. [ABSTRACT FROM AUTHOR]

Even though deep learning shows impressive results in several applications, its use on problems with High Dimensions and Low Sample Size, such as diagnosing rare diseases, leads to overfitting. One solution often proposed is feature selection. In deep learning, along with feature selection, network sparsification is also used to improve the results when dealing with high dimensions low sample size data. However, most of the time, they are tackled as separate problems. This paper proposes a new approach that integrates feature selection, based on sparsification, into the training process of a deep neural network. This approach uses a constrained biobjective gradient descent method. It provides a set of Pareto optimal neural networks that make a trade-off between network sparsity and model accuracy. Results on both artificial and real datasets show that using a constrained biobjective gradient descent increases the network sparsity without degrading the classification performances. With the proposed approach, on an artificial dataset, the feature selection score reached 0.97 with a sparsity score of 0.92 with an accuracy of 0.9. For the same accuracy, none of the other methods reached a feature score above 0.20 and sparsity score of 0.35. Finally, statistical tests validate the results obtained on all datasets. [ABSTRACT FROM AUTHOR]