Search

Railway transportation has been integrated into people’s lives. According to the “Notice on the release of the General Technical Specification of High-speed Railway Power Supply Safety Testing (6C System) System” issued by the National Railway Administration of China in 2012, it is required to install pantograph and slide monitoring devices in high-speed railway stations, station throats and the inlet and exit lines of high-speed railway sections, and it is required to detect the damage of the slider with high precision. It can be seen that the good condition of the pantograph slider is very important for the normal operation of the railway system. As a part of providing power for high-speed rail and subway, the pantograph must be paid attention to in railway transportation to ensure its integrity. The wear of the pantograph is mainly due to the contact power supply between the slide block and the long wire during high-speed operation, which inevitably produces scratches, resulting in depressions on the upper surface of the pantograph slide block. During long-term use, because the depression is too deep, there is a risk of fracture. Therefore, it is necessary to monitor the slider regularly and replace the slider with serious wear. At present, most of the traditional methods use automation technology or simple computer vision technology for detection, which is inefficient. Therefore, this paper introduces computer vision and deep learning technology into pantograph slide wear detection. Specifically, this paper mainly studies the wear detection of the pantograph slider based on deep learning and the main purpose is to improve the detection accuracy and improve the effect of segmentation. From a methodological perspective, this paper employs a linear array camera to enhance the quality of the data sets. Additionally, it integrates an attention mechanism to improve segmentation performance. Furthermore, this study introduces a novel image stitching method to address issues related to incomplete images, thereby providing a comprehensive solution.

Tool condition monitoring technology is an indispensable part of intelligent manufacturing. Most current research focuses on complex signal processing techniques or advanced deep learning algorithms to improve prediction performance without fully leveraging the end-to-end advantages of deep learning. The challenge lies in transforming multi-sensor raw data into input data suitable for direct model feeding, all while minimizing data scale and preserving sufficient temporal interpretation of tool wear. However, there is no clear reference standard for this so far. In light of this, this paper innovatively explores the processing methods that transform raw data into input data for deep learning models, a process known as an input paradigm. This paper introduces three new input paradigms: the downsampling paradigm, the periodic paradigm, and the subsequence paradigm. Then an improved hybrid model that combines a convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) was employed to validate the model’s performance. The subsequence paradigm demonstrated considerable superiority in prediction results based on the PHM2010 dataset, as the newly generated time series maintained the integrity of the raw data. Further investigation revealed that, with 120 subsequences and the temporal indicator being the maximum value, the model’s mean absolute error (MAE) and root mean square error (RMSE) were the lowest after threefold cross-validation, outperforming several classical and contemporary methods. The methods explored in this paper provide references for designing input data for deep learning models, helping to enhance the end-to-end potential of deep learning models, and promoting the industrial deployment and practical application of tool condition monitoring systems.

Understanding road conditions is essential for implementing effective road safety measures and driving solutions. Road situations encompass the day-to-day conditions of roads, including the presence of vehicles and pedestrians. Surveillance cameras strategically placed along streets have been instrumental in monitoring road situations and providing valuable information on pedestrians, moving vehicles, and objects within road environments. However, these video data and information are stored in large volumes, making analysis tedious and time-consuming. Deep learning models are increasingly utilized to monitor vehicles and identify and evaluate road and driving comfort situations. However, the current neural network model requires the recognition of situations using time-series video data. In this paper, we introduced a multi-directional detection model for road situations to uphold high accuracy. Deep learning methods often integrate long short-term memory (LSTM) into long-term recurrent network architectures. This approach effectively combines recurrent neural networks to capture temporal dependencies and convolutional neural networks (CNNs) to extract features from extensive video data. In our proposed method, we form a multi-directional long-term recurrent convolutional network approach with two groups equipped with CNN and two layers of LSTM. Additionally, we compare road situation recognition using convolutional neural networks, long short-term networks, and long-term recurrent convolutional networks. The paper presents a method for detecting and recognizing multi-directional road contexts using a modified LRCN. After balancing the dataset through data augmentation, the number of video files increased, resulting in our model achieving 91% accuracy, a significant improvement from the original dataset.

Brain stroke, or a cerebrovascular accident, is a devastating medical condition that disrupts the blood supply to the brain, depriving it of oxygen and nutrients. Each year, according to the World Health Organization, 15 million people worldwide experience a stroke. This results in approximately 5 million deaths and another 5 million individuals suffering permanent disabilities. The complex interplay of various risk factors highlights the urgent need for sophisticated analytical methods to more accurately predict stroke risks and manage their outcomes. Machine learning and deep learning technologies offer promising solutions by analyzing extensive datasets including patient demographics, health records, and lifestyle choices to uncover patterns and predictors not easily discernible by humans. These technologies enable advanced data processing, analysis, and fusion techniques for a comprehensive health assessment. We conducted a comprehensive review of 25 review papers published between 2020 and 2024 on machine learning and deep learning applications in brain stroke diagnosis, focusing on classification, segmentation, and object detection. Furthermore, all these reviews explore the performance evaluation and validation of advanced sensor systems in these areas, enhancing predictive health monitoring and personalized care recommendations. Moreover, we also provide a collection of the most relevant datasets used in brain stroke analysis. The selection of the papers was conducted according to PRISMA guidelines. Furthermore, this review critically examines each domain, identifies current challenges, and proposes future research directions, emphasizing the potential of AI methods in transforming health monitoring and patient care.

The paper “Using Absorption Models for Insulin and Carbohydrates and Deep Leaning to Improve Glucose Level Predictions” (Sensors 2021, 21, 5273) proposes a novel approach to predicting blood glucose levels for people with type 1 diabetes mellitus (T1DM). By building exponential models from raw carbohydrate and insulin data to simulate the absorption in the body, the authors reported a reduction in their model’s root-mean-square error (RMSE) from 15.5 mg/dL (raw) to 9.2 mg/dL (exponential) when predicting blood glucose levels one hour into the future. In this comment, we demonstrate that the experimental techniques used in that paper are flawed, which invalidates its results and conclusions. Specifically, after reviewing the authors’ code, we found that the model validation scheme was malformed, namely, the training and test data from the same time intervals were mixed. This means that the reported RMSE numbers in the referenced paper did not accurately measure the predictive capabilities of the approaches that were presented. We repaired the measurement technique by appropriately isolating the training and test data, and we discovered that their models actually performed dramatically worse than was reported in the paper. In fact, the models presented in the that paper do not appear to perform any better than a naive model that predicts future glucose levels to be the same as the current ones.

LoRa enables long-range communication for Internet of Things (IoT) devices, especially those with limited resources and low power requirements. Consequently, LoRa has emerged as a popular choice for numerous IoT applications. However, the security of LoRa devices is one of the major concerns that requires attention. Existing device identification mechanisms use cryptography which has two major issues: (1) cryptography is hard on the device resources and (2) physical attacks might prevent them from being effective. Deep learning-based radio frequency fingerprinting identification (RFFI) is emerging as a key candidate for device identification using hardware-intrinsic features. In this paper, we present a comprehensive survey of the state of the art in the area of deep learning-based radio frequency fingerprinting identification for LoRa devices. We discuss various categories of radio frequency fingerprinting techniques along with hardware imperfections that can be exploited to identify an emitter. Furthermore, we describe different deep learning algorithms implemented for the task of LoRa device classification and summarize the main approaches and results. We discuss several representations of the LoRa signal used as input to deep learning models. Additionally, we provide a thorough review of all the LoRa RF signal datasets used in the literature and summarize details about the hardware used, the type of signals collected, the features provided, availability, and size. Finally, we conclude this paper by discussing the existing challenges in deep learning-based LoRa device identification and also envisage future research directions and opportunities.

Unmanned aerial vehicle (UAV)-based object detection methods are widely used in traffic detection due to their high flexibility and extensive coverage. In recent years, with the increasing complexity of the urban road environment, UAV object detection algorithms based on deep learning have gradually become a research hotspot. However, how to further improve algorithmic efficiency in response to the numerous and rapidly changing road elements, and thus achieve high-speed and accurate road object detection, remains a challenging issue. Given this context, this paper proposes the high-efficiency multi-object detection algorithm for UAVs (HeMoDU). HeMoDU reconstructs a state-of-the-art, deep-learning-based object detection model and optimizes several aspects to improve computational efficiency and detection accuracy. To validate the performance of HeMoDU in urban road environments, this paper uses the public urban road datasets VisDrone2019 and UA-DETRAC for evaluation. The experimental results show that the HeMoDU model effectively improves the speed and accuracy of UAV object detection.

Scene text detection is an important research field in computer vision, playing a crucial role in various application scenarios. However, existing scene text detection methods often fail to achieve satisfactory results when faced with text instances of different sizes, shapes, and complex backgrounds. To address the challenge of detecting diverse texts in natural scenes, this paper proposes a multi-scale natural scene text detection method based on attention feature extraction and cascaded feature fusion. This method combines global and local attention through an improved attention feature fusion module (DSAF) to capture text features of different scales, enhancing the network’s perception of text regions and improving its feature extraction capabilities. Simultaneously, an improved cascaded feature fusion module (PFFM) is used to fully integrate the extracted feature maps, expanding the receptive field of features and enriching the expressive ability of the feature maps. Finally, to address the cascaded feature maps, a lightweight subspace attention module (SAM) is introduced to partition the concatenated feature maps into several sub-space feature maps, facilitating spatial information interaction among features of different scales. In this paper, comparative experiments are conducted on the ICDAR2015, Total-Text, and MSRA-TD500 datasets, and comparisons are made with some existing scene text detection methods. The results show that the proposed method achieves good performance in terms of accuracy, recall, and F-score, thus verifying its effectiveness and practicality.

Sign language is an essential means of communication for individuals with hearing disabilities. However, there is a significant shortage of sign language interpreters in some languages, especially in Saudi Arabia. This shortage results in a large proportion of the hearing-impaired population being deprived of services, especially in public places. This paper aims to address this gap in accessibility by leveraging technology to develop systems capable of recognizing Arabic Sign Language (ArSL) using deep learning techniques. In this paper, we propose a hybrid model to capture the spatio-temporal aspects of sign language (i.e., letters and words). The hybrid model consists of a Convolutional Neural Network (CNN) classifier to extract spatial features from sign language data and a Long Short-Term Memory (LSTM) classifier to extract spatial and temporal characteristics to handle sequential data (i.e., hand movements). To demonstrate the feasibility of our proposed hybrid model, we created a dataset of 20 different words, resulting in 4000 images for ArSL: 10 static gesture words and 500 videos for 10 dynamic gesture words. Our proposed hybrid model demonstrates promising performance, with the CNN and LSTM classifiers achieving accuracy rates of 94.40% and 82.70%, respectively. These results indicate that our approach can significantly enhance communication accessibility for the hearing-impaired community in Saudi Arabia. Thus, this paper represents a major step toward promoting inclusivity and improving the quality of life for the hearing impaired.

Visual simultaneous localization and mapping (VSLAM) enhances the navigation of autonomous agents in unfamiliar environments by progressively constructing maps and estimating poses. However, conventional VSLAM pipelines often exhibited degraded performance in dynamic environments featuring mobile objects. Recent research in deep learning led to notable progress in semantic segmentation, which involves assigning semantic labels to image pixels. The integration of semantic segmentation into VSLAM can effectively differentiate between static and dynamic elements in intricate scenes. This paper provided a comprehensive comparative review on leveraging semantic segmentation to improve major components of VSLAM, including visual odometry, loop closure detection, and environmental mapping. Key principles and methods for both traditional VSLAM and deep semantic segmentation were introduced. This paper presented an overview and comparative analysis of the technical implementations of semantic integration across various modules of the VSLAM pipeline. Furthermore, it examined the features and potential use cases associated with the fusion of VSLAM and semantics. It was found that the existing VSLAM model continued to face challenges related to computational complexity. Promising future research directions were identified, including efficient model design, multimodal fusion, online adaptation, dynamic scene reconstruction, and end-to-end joint optimization. This review shed light on the emerging paradigm of semantic VSLAM and how deep learning-enabled semantic reasoning could unlock new capabilities for autonomous intelligent systems to operate reliably in the real world.

This paper proposes a multimodal Transformer model that uses time-series data to detect and predict winter road surface conditions. For detecting or predicting road surface conditions, the previous approach focuses on the cooperative use of multiple modalities as inputs, e.g., images captured by fixed-point cameras (road surface images) and auxiliary data related to road surface conditions under simple modality integration. Although such an approach achieves performance improvement compared to the method using only images or auxiliary data, there is a demand for further consideration of the way to integrate heterogeneous modalities. The proposed method realizes a more effective modality integration using a cross-attention mechanism and time-series processing. Concretely, when integrating multiple modalities, feature compensation through mutual complementation between modalities is realized through a feature integration technique based on a cross-attention mechanism, and the representational ability of the integrated features is enhanced. In addition, by introducing time-series processing for the input data across several timesteps, it is possible to consider the temporal changes in the road surface conditions. Experiments are conducted for both detection and prediction tasks using data corresponding to the current winter condition and data corresponding to a few hours after the current winter condition, respectively. The experimental results verify the effectiveness of the proposed method for both tasks. In addition to the construction of the classification model for winter road surface conditions, we first attempt to visualize the classification results, especially the prediction results, through the image style transfer model as supplemental extended experiments on image generation at the end of the paper.

Emotion recognition has become increasingly important in the field of Deep Learning (DL) and computer vision due to its broad applicability by using human–computer interaction (HCI) in areas such as psychology, healthcare, and entertainment. In this paper, we conduct a systematic review of facial and pose emotion recognition using DL and computer vision, analyzing and evaluating 77 papers from different sources under Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Our review covers several topics, including the scope and purpose of the studies, the methods employed, and the used datasets. The scope of this work is to conduct a systematic review of facial and pose emotion recognition using DL methods and computer vision. The studies were categorized based on a proposed taxonomy that describes the type of expressions used for emotion detection, the testing environment, the currently relevant DL methods, and the datasets used. The taxonomy of methods in our review includes Convolutional Neural Network (CNN), Faster Region-based Convolutional Neural Network (R-CNN), Vision Transformer (ViT), and “Other NNs”, which are the most commonly used models in the analyzed studies, indicating their trendiness in the field. Hybrid and augmented models are not explicitly categorized within this taxonomy, but they are still important to the field. This review offers an understanding of state-of-the-art computer vision algorithms and datasets for emotion recognition through facial expressions and body poses, allowing researchers to understand its fundamental components and trends.

With the increasing complexity of the grid meter dial, precise feature extraction is becoming more and more difficult. Many automatic recognition solutions have been proposed for grid meter readings. However, traditional inspection methods cannot guarantee detection accuracy in complex environments. So, deep-learning methods are combined with grid meter recognition. Existing recognition systems that utilize segmentation models exhibit very high computation. It is challenging to ensure high real-time performance in edge computing devices. Therefore, an improved meter recognition model based on YOLOv7 is proposed in this paper. Partial convolution (PConv) is introduced into YOLOv7 to create a lighter network. Different PConv introduction locations on the base module have been used in order to find the optimal approach for reducing the parameters and floating point of operations (FLOPs). Meanwhile, the dynamic head (DyHead) module is utilized to enhance the attention mechanism for the YOLOv7 model. It can improve the detection accuracy of striped objects. As a result, this paper achieves mAP50val of 97.87% and mAP50:90val of 62.4% with only 5.37 M parameters. The improved model’s inference speed can reach 108 frames per second (FPS). It enables detection accuracy that can reach ±0.1 degrees in the grid meter.

In response to the numerous challenges faced by traditional human pose recognition methods in practical applications, such as dense targets, severe edge occlusion, limited application scenarios, complex backgrounds, and poor recognition accuracy when targets are occluded, this paper proposes a YOLO-Pose algorithm for human pose estimation. The specific improvements are divided into four parts. Firstly, in the Backbone section of the YOLO-Pose model, lightweight GhostNet modules are introduced to reduce the model’s parameter count and computational requirements, making it suitable for deployment on unmanned aerial vehicles (UAVs). Secondly, the ACmix attention mechanism is integrated into the Neck section to improve detection speed during object judgment and localization. Furthermore, in the Head section, key points are optimized using coordinate attention mechanisms, significantly enhancing key point localization accuracy. Lastly, the paper improves the loss function and confidence function to enhance the model’s robustness. Experimental results demonstrate that the improved model achieves a 95.58% improvement in mAP50 and a 69.54% improvement in mAP50-95 compared to the original model, with a reduction of 14.6 M parameters. The model achieves a detection speed of 19.9 ms per image, optimized by 30% and 39.5% compared to the original model. Comparisons with other algorithms such as Faster R-CNN, SSD, YOLOv4, and YOLOv7 demonstrate varying degrees of performance improvement.

This paper investigates a method for precise mapping of human arm movements using sEMG signals. A multi-channel approach captures the sEMG signals, which, combined with the accurately calculated joint angles from an Inertial Measurement Unit, allows for action recognition and mapping through deep learning algorithms. Firstly, signal acquisition and processing were carried out, which involved acquiring data from various movements (hand gestures, single-degree-of-freedom joint movements, and continuous joint actions) and sensor placement. Then, interference signals were filtered out through filters, and the signals were preprocessed using normalization and moving averages to obtain sEMG signals with obvious features. Additionally, this paper constructs a hybrid network model, combining Convolutional Neural Networks and Artificial Neural Networks, and employs a multi-feature fusion algorithm to enhance the accuracy of gesture recognition. Furthermore, a nonlinear fitting between sEMG signals and joint angles was established based on a backpropagation neural network, incorporating momentum term and adaptive learning rate adjustments. Finally, based on the gesture recognition and joint angle prediction model, prosthetic arm control experiments were conducted, achieving highly accurate arm movement prediction and execution. This paper not only validates the potential application of sEMG signals in the precise control of robotic arms but also lays a solid foundation for the development of more intuitive and responsive prostheses and assistive devices.

Deep learning (DL) has been frequently applied in the image reconstruction of electromagnetic tomography (EMT) in recent years. It offers the potential to achieve higher-quality image reconstruction. Among these, research on samples is relatively scarce. Samples are the cornerstone for both large and small models, which is easy to ignore. In this paper, a deep learning electromagnetic tomography (DL-EMT) model with nine elements is established. Complete simulation and experimental samples are obtained based on this model. On the sample sets, the reconstruction quality is observed by adjusting the size and configuration of the training set. The Mann–Whitney U test shows that beyond a certain point, the addition of more samples to the training data fed into the deep learning network does not result in an obvious improvement statistically in the quality of the reconstructed images. This paper proposes a CC-building method for optimizing a sample set. This method is based on the Pearson correlation coefficient calculation, aiming to establish a more effective sample base for DL-EMT image reconstruction. The statistical analysis shows that the CC-building method can significantly improve the image reconstruction effect in a small and moderate sample size. This method is also validated by experiments.

Mobile crowdsensing (MCS) systems rely on the collective contribution of sensor data from numerous mobile devices carried by participants. However, the open and participatory nature of MCS renders these systems vulnerable to adversarial attacks or data poisoning attempts where threat actors can inject malicious data into the system. There is a need for a detection system that mitigates malicious sensor data to maintain the integrity and reliability of the collected information. This paper addresses this issue by proposing an adaptive and robust model for detecting malicious data in MCS scenarios involving sensor data from mobile devices. The proposed model incorporates an adaptive learning mechanism that enables the TCN-based model to continually evolve and adapt to new patterns, enhancing its capability to detect novel malicious data as threats evolve. We also present a comprehensive evaluation of the proposed model’s performance using the SherLock datasets, demonstrating its effectiveness in accurately detecting malicious sensor data and mitigating potential threats to the integrity of MCS systems. Comparative analysis with existing models highlights the performance of the proposed TCN-based model in terms of detection accuracy, with an accuracy score of 98%. Through these contributions, the paper aims to advance the state of the art in ensuring the trustworthiness and security of MCS systems, paving the way for the development of more reliable and robust crowdsensing applications.

The cooperative, connected, and automated mobility (CCAM) infrastructure plays a key role in understanding and enhancing the environmental perception of autonomous vehicles (AVs) driving in complex urban settings. However, the deployment of CCAM infrastructure necessitates the efficient selection of the computational processing layer and deployment of machine learning (ML) and deep learning (DL) models to achieve greater performance of AVs in complex urban environments. In this paper, we propose a computational framework and analyze the effectiveness of a custom-trained DL model (YOLOv8) when deployed in diverse devices and settings at the vehicle-edge-cloud-layered architecture. Our main focus is to understand the interplay and relationship between the DL model’s accuracy and execution time during deployment at the layered framework. Therefore, we investigate the trade-offs between accuracy and time by the deployment process of the YOLOv8 model over each layer of the computational framework. We consider the CCAM infrastructures, i.e., sensory devices, computation, and communication at each layer. The findings reveal that the performance metrics results (e.g., 0.842 mAP@0.5) of deployed DL models remain consistent regardless of the device type across any layer of the framework. However, we observe that inference times for object detection tasks tend to decrease when the DL model is subjected to different environmental conditions. For instance, the Jetson AGX (non-GPU) outperforms the Raspberry Pi (non-GPU) by reducing inference time by 72%, whereas the Jetson AGX Xavier (GPU) outperforms the Jetson AGX ARMv8 (non-GPU) by reducing inference time by 90%. A complete average time comparison analysis for the transfer time, preprocess time, and total time of devices Apple M2 Max, Intel Xeon, Tesla T4, NVIDIA A100, Tesla V100, etc., is provided in the paper. Our findings direct the researchers and practitioners to select the most appropriate device type and environment for the deployment of DL models required for production.

In order to effectively respond to floods and water emergencies that result in the drowning of missing persons, timely and effective search and rescue is a very critical step in underwater rescue. Due to the complex underwater environment and low visibility, unmanned underwater vehicles (UUVs) with sonar are more efficient than traditional manual search and rescue methods to conduct active searches using deep learning algorithms. In this paper, we constructed a sound-based rescue target dataset that encompasses both the source and target domains using deep transfer learning techniques. For the underwater acoustic rescue target detection of small targets, which lack image feature accuracy, this paper proposes a two-branch convolution module and improves the YOLOv5s algorithm model to design an acoustic rescue small target detection algorithm model. For an underwater rescue target dataset based on acoustic images with a small sample acoustic dataset, a direct fine-tuning using optical image pre-training lacks cross-domain adaptability due to the different statistical properties of optical and acoustic images. This paper therefore proposes a heterogeneous information hierarchical migration learning method. For the false detection of acoustic rescue targets in a complex underwater background, the network layer is frozen during the hierarchical migration of heterogeneous information to improve the detection accuracy. In addition, in order to be more applicable to the embedded devices carried by underwater UAVs, an underwater acoustic rescue target detection algorithm based on ShuffleNetv2 is proposed to improve the two-branch convolutional module and the backbone network of YOLOv5s algorithm, and to create a lightweight model based on hierarchical migration of heterogeneous information. Through extensive comparative experiments conducted on various acoustic images, we have thoroughly validated the feasibility and effectiveness of our method. Our approach has demonstrated state-of-the-art performance in underwater search and rescue target detection tasks.

This paper proposes a new approach to defect detection system design focused on exact damaged areas demonstrated through visual data containing gear wheel images. The main advantage of the system is the capability to detect a wide range of patterns of defects occurring in datasets. The methodology is built on three processes that combine different approaches from unsupervised and supervised methods. The first step is a search for anomalies, which is performed by defining the correct areas on the controlled object by using the autoencoder approach. As a result, the differences between the original and autoencoder-generated images are obtained. These are divided into clusters using the clustering method (DBSCAN). Based on the clusters, the regions of interest are subsequently defined and classified using the pre-trained Xception network classifier. The main result is a system capable of focusing on exact defect areas using the sequence of unsupervised learning (autoencoder)–unsupervised learning (clustering)–supervised learning (classification) methods (U2S-CNN). The outcome with tested samples was 177 detected regions and 205 occurring damaged areas. There were 108 regions detected correctly, and 69 regions were labeled incorrectly. This paper describes a proof of concept for defect detection by highlighting exact defect areas. It can be thus an alternative to using detectors such as YOLO methods, reconstructors, autoencoders, transformers, etc.

For the synthesis of ultra-large scene and ultra-high resolution videos, in order to obtain high-quality large-scene videos, high-quality video stitching and fusion are achieved through multi-scale unstructured array cameras. This paper proposes a network model image feature point extraction algorithm based on symmetric auto-encoding and scale feature fusion. By using the principle of symmetric auto-encoding, the hierarchical restoration of image feature location information is incorporated into the corresponding scale feature, along with deep separable convolution image feature extraction, which not only improves the performance of feature point detection but also significantly reduces the computational complexity of the network model. Based on the calculated high-precision feature point pairing information, a new image localization method is proposed based on area ratio and homography matrix scaling, which improves the speed and accuracy of the array camera image scale alignment and positioning, realizes high-definition perception of local details in large scenes, and obtains clearer synthesis effects of large scenes and high-quality stitched images. The experimental results show that the feature point extraction algorithm proposed in this paper has been experimentally compared with four typical algorithms using the HPatches dataset. The performance of feature point detection has been improved by an average of 4.9%, the performance of homography estimation has been improved by an average of 2.5%, the amount of computation has been reduced by 18%, the number of network model parameters has been reduced by 47%, and the synthesis of billion-pixel videos has been achieved, demonstrating practicality and robustness.

This paper provides a comprehensive overview of affective computing systems for facial expression recognition (FER) research in naturalistic contexts. The first section presents an updated account of user-friendly FER toolboxes incorporating state-of-the-art deep learning models and elaborates on their neural architectures, datasets, and performances across domains. These sophisticated FER toolboxes can robustly address a variety of challenges encountered in the wild such as variations in illumination and head pose, which may otherwise impact recognition accuracy. The second section of this paper discusses multimodal large language models (MLLMs) and their potential applications in affective science. MLLMs exhibit human-level capabilities for FER and enable the quantification of various contextual variables to provide context-aware emotion inferences. These advancements have the potential to revolutionize current methodological approaches for studying the contextual influences on emotions, leading to the development of contextualized emotion models.

A massive number of paper documents that include important information such as circuit schematics can be converted into digital documents by optical sensors like scanners or digital cameras. However, extracting the netlists of analog circuits from digital documents is an exceptionally challenging task. This process aids enterprises in digitizing paper-based circuit diagrams, enabling the reuse of analog circuit designs and the automatic generation of datasets required for intelligent design models in this domain. This paper introduces a bottom-up graph encoding model aimed at automatically parsing the circuit topology of analog integrated circuits from images. The model comprises an improved electronic component detection network based on the Swin Transformer, an algorithm for component port localization, and a graph encoding model. The objective of the detection network is to accurately identify component positions and types, followed by automatic dataset generation through port localization, and finally, utilizing the graph encoding model to predict potential connections between circuit components. To validate the model’s performance, we annotated an electronic component detection dataset and a circuit diagram dataset, comprising 1200 and 3552 training samples, respectively. Detailed experimentation results demonstrate the superiority of our proposed enhanced algorithm over comparative algorithms across custom and public datasets. Furthermore, our proposed port localization algorithm significantly accelerates the annotation speed of circuit diagram datasets.

Advanced driver assistance systems (ADASs) are becoming increasingly common in modern-day vehicles, as they not only improve safety and reduce accidents but also aid in smoother and easier driving. ADASs rely on a variety of sensors such as cameras, radars, lidars, and a combination of sensors, to perceive their surroundings and identify and track objects on the road. The key components of ADASs are object detection, recognition, and tracking algorithms that allow vehicles to identify and track other objects on the road, such as other vehicles, pedestrians, cyclists, obstacles, traffic signs, traffic lights, etc. This information is then used to warn the driver of potential hazards or used by the ADAS itself to take corrective actions to avoid an accident. This paper provides a review of prominent state-of-the-art object detection, recognition, and tracking algorithms used in different functionalities of ADASs. The paper begins by introducing the history and fundamentals of ADASs followed by reviewing recent trends in various ADAS algorithms and their functionalities, along with the datasets employed. The paper concludes by discussing the future of object detection, recognition, and tracking algorithms for ADASs. The paper also discusses the need for more research on object detection, recognition, and tracking in challenging environments, such as those with low visibility or high traffic density.

This paper proposes a hybrid deep learning emotion classification system (HDECS), a hybrid multimodal deep learning system designed for emotion classification in a specific national language. Emotion classification is important in diverse fields, including tailored corporate services, AI advancement, and more. Additionally, most sentiment classification techniques in speaking situations are based on a single modality: voice, conversational text, vital signs, etc. However, analyzing these data presents challenges because of the variations in vocal intonation, text structures, and the impact of external stimuli on physiological signals. Korean poses challenges in natural language processing, including subject omission and spacing issues. To overcome these challenges and enhance emotion classification performance, this paper presents a case study using Korean multimodal data. The case study model involves retraining two pretrained models, LSTM and CNN, until their predictions on the entire dataset reach an agreement rate exceeding 0.75. Predictions are used to generate emotional sentences appended to script data, which are further processed using BERT for final emotion prediction. The research result is evaluated by using categorical cross-entropy (CCE) to measure the difference between the model’s predictions and actual labels, F1 score, and accuracy. According to the evaluation, the case model outperforms the existing KLUE/roBERTa model with improvements of 0.5 in CCE, 0.09 in accuracy, and 0.11 in F1 score. As a result, the HDECS is expected to perform well not only on Korean multimodal datasets but also on sentiment classification considering the speech characteristics of various languages and regions.

Tunnel cracks are the main factors that cause damage and collapse of tunnel structures. How to detect tunnel cracks efficiently and avoid safety accidents caused by tunnel cracks effectively is a research hotspot at present. In order to meet the need for efficient detection of tunnel cracks, the tunnel crack detection method based on improved Retinex and deep learning is proposed in this paper. The tunnel crack images collected by optical imaging equipment are used to improve the contrast information of tunnel crack images using the image enhancement algorithm, and this image enhancement algorithm has the function of multi-scale Retinex decomposition with improved central filtering. An improved VGG19 network model is constructed to achieve efficient segmentation of tunnel crack images through deep learning methods and then form the segmented binary image. The Zhang–Suen fast parallel-thinning method is used to obtain the skeleton map of the single-layer pixel, and the length and width information of the tunnel cracks are obtained. The feasibility and effectiveness of the proposed method are verified by experiments. Compared with other methods in the literature, the maximum deviation in the length of the tunnel crack is about 5 mm, and the maximum deviation in the width of the tunnel crack is about 0.8 mm. The experimental results show that the proposed method has a shorter detection time and higher detection accuracy. The research results of this paper can provide a strong basis for the health evaluation of tunnels.

An autonomous place recognition system is essential for scenarios where GPS is useless, such as underground tunnels. However, it is difficult to use existing algorithms to fully utilize the small number of effective features in underground tunnel data, and recognition accuracy is difficult to guarantee. In order to solve this challenge, an efficient point cloud position recognition algorithm, named Dual-Attention Transformer Network (DAT-Net), is proposed in this paper. The algorithm firstly adopts the farthest point downsampling module to eliminate the invalid redundant points in the point cloud data and retain the basic shape of the point cloud, which reduces the size of the point cloud and, at the same time, reduces the influence of the invalid point cloud on the data analysis. After that, this paper proposes the dual-attention Transformer module to facilitate local information exchange by utilizing the multi-head self-attention mechanism. It extracts local descriptors and integrates highly discriminative global descriptors based on global context with the help of a feature fusion layer to obtain a more accurate and robust global feature representation. Experimental results show that the method proposed in this paper achieves an average F1 score of 0.841 on the SubT-Tunnel dataset and outperforms many existing state-of-the-art algorithms in recognition accuracy and robustness tests.

Vehicle ad hoc networks (VANETs) are a vital part of intelligent transportation systems (ITS), offering a variety of advantages from reduced traffic to increased road safety. Despite their benefits, VANETs remain vulnerable to various security threats, including severe blackhole attacks. In this paper, we propose a deep-learning-based secure routing (DLSR) protocol using a deep-learning-based clustering (DLC) protocol to establish a secure route against blackhole attacks. The main features and contributions of this paper are as follows. First, the DLSR protocol utilizes deep learning (DL) at each node to choose secure routing or normal routing while establishing secure routes. Additionally, we can identify the behavior of malicious nodes to determine the best possible next hop based on its fitness function value. Second, the DLC protocol is considered an underlying structure to enhance connectivity between nodes and reduce control overhead. Third, we design a deep neural network (DNN) model to optimize the fitness function in both DLSR and DLC protocols. The DLSR protocol considers parameters such as remaining energy, distance, and hop count, while the DLC protocol considers cosine similarity, cosine distance, and the node’s remaining energy. Finally, from the performance results, we evaluate the performance of the proposed routing and clustering protocol in the viewpoints of packet delivery ratio, routing delay, control overhead, packet loss ratio, and number of packet losses. Additionally, we also exploit the impact of the mobility model such as reference point group mobility (RPGM) and random waypoint (RWP) on the network metrics.

Colony-Forming Unit (CFU) counting is a complex problem without a universal solution in biomedical and food safety domains. A multitude of sophisticated heuristics and segmentation-driven approaches have been proposed by researchers. However, U-Net remains the most frequently cited and used deep learning method in these domains. The latter approach provides a segmentation output map and requires an additional counting procedure to calculate unique segmented regions and detect microbial colonies. However, due to pixel-based targets, it tends to generate irrelevant artifacts or errant pixels, leading to inaccurate and mixed post-processing results. In response to these challenges, this paper proposes a novel hybrid counting approach, incorporating a multi-loss U-Net reformulation and a post-processing Petri dish localization algorithm. Firstly, a unique innovation lies in the multi-loss U-Net reformulation. An additional loss term is introduced in the bottleneck U-Net layer, focusing on the delivery of an auxiliary signal that indicates where to look for distinct CFUs. Secondly, the novel localization algorithm automatically incorporates an agar plate and its bezel into the CFU counting techniques. Finally, the proposition is further enhanced by the integration of a fully automated solution, which comprises a specially designed uniform Petri dish illumination system and a counting web application. The latter application directly receives images from the camera, processes them, and sends the segmentation results to the user. This feature provides an opportunity to correct the CFU counts, offering a feedback loop that contributes to the continued development of the deep learning model. Through extensive experimentation, the authors of this paper have found that all probed multi-loss U-Net architectures incorporated into the proposed hybrid approach consistently outperformed their single-loss counterparts, as well as other comparable models such as self-normalized density maps and YOLOv6, by at least 1% to 3% in mean absolute and symmetric mean absolute percentage errors. Further significant improvements were also reported through the means of the novel localization algorithm. This reaffirms the effectiveness of the proposed hybrid solution in addressing contemporary challenges of precise in vitro CFU counting.

Aiming at the problems of the poor recognition effect and low recognition rate of the existing methods in the process of belt deviation detection, this paper proposes a real-time belt deviation detection method. Firstly, ResNet18 combined with the attention mechanism module is used as a feature extraction network to enhance the features in the belt edge region and suppress the features in other regions. Then, the extracted features are used to predict the approximate locations of the belt edges using a classifier based on the contextual information on the fully connected layer. Next, the improved gradient equation is used as a structural loss in the model training stage to make the model prediction value closer to the target value. Then, the authors of this paper use the least squares method to fit the set of detected belt edge line points to obtain the accurate belt edge straight line. Finally, the deviation threshold is set according to the requirements of the safety production code, and the fitting results are compared with the threshold to achieve the belt deviation detection. Comparisons are made with four other methods: ultrafast structure-aware deep lane detection, end-to-end wireframe parsing, LSD, and the Hough transform. The results show that the proposed method is the fastest at 41 frames/sec; the accuracy is improved by 0.4%, 13.9%, 45.9%, and 78.8% compared to the other four methods; and the F1-score index is improved by 0.3%, 10.2%, 32.6%, and 72%, respectively, which meets the requirements of practical engineering applications. The proposed method can be used for intelligent monitoring and control in coal mines, logistics and transport industries, and other scenarios requiring belt transport.

The network area is extended from ground to air. In order to efficiently manage various kinds of nodes, new network paradigms are needed such as cell-free massive multiple-input multiple-output (CF-mMIMO). Additionally, security is also considered as one of the important quality-of-services (QoS) parameters in future networks. Thus, in this paper, we propose a novel deep learning-based secure multicast routing protocol (DLSMR) in flying ad hoc networks (FANETs) with cell-free massive MIMO (CF-mMIMO). We consider the problem of wormhole attacks in the multicast routing process. To tackle this problem, we propose the DLSMR protocol, which utilizes a deep learning (DL) approach to predict the secure and unsecured route based on node ID, distance, destination sequence, hop count, and energy to avoid wormhole attacks. This work also addresses key concerns in FANETs such as security, scalability, and stability. The main contributions of this paper are as follows: (1) We propose a deep learning-based secure multicast routing protocol (DLSMR) to establish a high-stability multicast tree and improve security performance against wormhole attacks. In more detail, the DLSMR protocol predicts whether the route is secure based on network information such as node ID, distance, destination sequence, hop count, and remaining energy or not. (2) To improve the node connectivity and manage multicast members, we propose a top-down particle swarm optimization-based clustering (TD-PSO) protocol to maximize the cost function considering node degree, cosine similarity, cosine distance, and cluster head energy to guarantee convergence to the global optima. Thus, the TD-PSO protocol provides more strong connectivity. (3) Performance evaluations verify the proposed routing protocol establishes a secure route by avoiding wormhole attacks as well as by providing strong connectivity. The TD-PSO clustering supports connectivity to enhance network performance. In addition, we exploit the impact of the mobility model on the network metrics such as packet delivery ratio, routing delay, control overhead, packet loss ratio, and number of packet losses.

The nine-axis inertial and measurement unit (IMU)-based three-dimensional (3D) orientation estimation is a fundamental part of inertial motion capture. Recently, owing to the successful utilization of deep learning in various applications, orientation estimation neural networks (NNs) trained on large datasets, including nine-axis IMU signals and reference orientation data, have been developed. During the training process, the limited amount of training data is a critical issue in the development of powerful networks. Data augmentation, which increases the amount of training data, is a key approach for addressing the data shortage problem and thus for improving the estimation performance. However, to the best of our knowledge, no studies have been conducted to analyze the effects of data augmentation techniques on estimation performance in orientation estimation networks using IMU sensors. This paper selects three data augmentation techniques for IMU-based orientation estimation NNs, i.e., augmentation by virtual rotation, bias addition, and noise addition (which are hereafter referred to as rotation, bias, and noise, respectively). Then, this paper analyzes the effects of these augmentation techniques on estimation accuracy in recurrent neural networks, for a total of seven combinations (i.e., rotation only, bias only, noise only, rotation and bias, rotation and noise, and rotation and bias and noise). The evaluation results show that, among a total of seven augmentation cases, four cases including ‘rotation’ (i.e., rotation only, rotation and bias, rotation and noise, and rotation and bias and noise) occupy the top four. Therefore, it may be concluded that the augmentation effect of rotation is overwhelming compared to those of bias and noise. By applying rotation augmentation, the performance of the NN can be significantly improved. The analysis of the effect of the data augmentation techniques presented in this paper may provide insights for developing robust IMU-based orientation estimation networks.

Gestures have been used for nonverbal communication for a long time, but human–computer interaction (HCI) via gestures is becoming more common in the modern era. To obtain a greater recognition rate, the traditional interface comprises various devices, such as gloves, physical controllers, and markers. This study provides a new markerless technique for obtaining gestures without the need for any barriers or pricey hardware. In this paper, dynamic gestures are first converted into frames. The noise is removed, and intensity is adjusted for feature extraction. The hand gesture is first detected through the images, and the skeleton is computed through mathematical computations. From the skeleton, the features are extracted; these features include joint color cloud, neural gas, and directional active model. After that, the features are optimized, and a selective feature set is passed through the classifier recurrent neural network (RNN) to obtain the classification results with higher accuracy. The proposed model is experimentally assessed and trained over three datasets: HaGRI, Egogesture, and Jester. The experimental results for the three datasets provided improved results based on classification, and the proposed system achieved an accuracy of 92.57% over HaGRI, 91.86% over Egogesture, and 91.57% over the Jester dataset, respectively. Also, to check the model liability, the proposed method was tested on the WLASL dataset, attaining 90.43% accuracy. This paper also includes a comparison with other-state-of-the art methods to compare our model with the standard methods of recognition. Our model presented a higher accuracy rate with a markerless approach to save money and time for classifying the gestures for better interaction.

Fire incidents occurring onboard ships cause significant consequences that result in substantial effects. Fires on ships can have extensive and severe wide-ranging impacts on matters such as the safety of the crew, cargo, the environment, finances, reputation, etc. Therefore, timely detection of fires is essential for quick responses and powerful mitigation. The study in this research paper presents a fire detection technique based on YOLOv7 (You Only Look Once version 7), incorporating improved deep learning algorithms. The YOLOv7 architecture, with an improved E-ELAN (extended efficient layer aggregation network) as its backbone, serves as the basis of our fire detection system. Its enhanced feature fusion technique makes it superior to all its predecessors. To train the model, we collected 4622 images of various ship scenarios and performed data augmentation techniques such as rotation, horizontal and vertical flips, and scaling. Our model, through rigorous evaluation, showcases enhanced capabilities of fire recognition to improve maritime safety. The proposed strategy successfully achieves an accuracy of 93% in detecting fires to minimize catastrophic incidents. Objects having visual similarities to fire may lead to false prediction and detection by the model, but this can be controlled by expanding the dataset. However, our model can be utilized as a real-time fire detector in challenging environments and for small-object detection. Advancements in deep learning models hold the potential to enhance safety measures, and our proposed model in this paper exhibits this potential. Experimental results proved that the proposed method can be used successfully for the protection of ships and in monitoring fires in ship port areas. Finally, we compared the performance of our method with those of recently reported fire-detection approaches employing widely used performance matrices to test the fire classification results achieved.

Facial emotion recognition (FER) is a computer vision process aimed at detecting and classifying human emotional expressions. FER systems are currently used in a vast range of applications from areas such as education, healthcare, or public safety; therefore, detection and recognition accuracies are very important. Similar to any computer vision task based on image analyses, FER solutions are also suitable for integration with artificial intelligence solutions represented by different neural network varieties, especially deep neural networks that have shown great potential in the last years due to their feature extraction capabilities and computational efficiency over large datasets. In this context, this paper reviews the latest developments in the FER area, with a focus on recent neural network models that implement specific facial image analysis algorithms to detect and recognize facial emotions. This paper’s scope is to present from historical and conceptual perspectives the evolution of the neural network architectures that proved significant results in the FER area. This paper endorses convolutional neural network (CNN)-based architectures against other neural network architectures, such as recurrent neural networks or generative adversarial networks, highlighting the key elements and performance of each architecture, and the advantages and limitations of the proposed models in the analyzed papers. Additionally, this paper presents the available datasets that are currently used for emotion recognition from facial expressions and micro-expressions. The usage of FER systems is also highlighted in various domains such as healthcare, education, security, or social IoT. Finally, open issues and future possible developments in the FER area are identified.

As the demand for Internet access increases, malicious traffic on the Internet has soared also. In view of the fact that the existing malicious-traffic-identification methods suffer from low accuracy, this paper proposes a malicious-traffic-identification method based on contrastive learning. The proposed method is able to overcome the shortcomings of traditional methods that rely on labeled samples and is able to learn data feature representations carrying semantic information from unlabeled data, thus improving the model accuracy. In this paper, a new malicious traffic feature extraction model based on a Transformer is proposed. Employing a self-attention mechanism, the proposed feature extraction model can extract the bytes features of malicious traffic by performing calculations on the malicious traffic, thereby realizing the efficient identification of malicious traffic. In addition, a bidirectional GLSTM is introduced to extract the timing features of malicious traffic. The experimental results show that the proposed method is superior to the latest published methods in terms of accuracy and F1 score.

In order to more effectively monitor and interfere with enemy signals, it is particularly important to accurately and efficiently identify the intercepted signals and estimate their parameters in the increasingly complex electromagnetic environment. Therefore, in non-cooperative situations, it is of great practical significance to study how to accurately detect direct sequence spread spectrum (DSSS) signals in real time and estimate their parameters. The traditional time-delay correlation algorithm encounters the challenges such as peak energy leakage and false peak interference. As an alternative, this paper introduces a Pseudo-Noise (PN) code period estimation method utilizing a one-dimensional (1D) convolutional neural network based on the residual network (CNN-ResNet). This method transforms the problem of spread spectrum code period estimation into a multi-classification problem of spread spectrum code length estimation. Firstly, the In-phase/Quadrature(I/Q) two-way of the received DSSS signals is directly input into the CNN-ResNet model, which will automatically learn the characteristics of the DSSS signal with different PN code lengths and then estimate the PN code length. Simulation experiments are conducted using a data set with DSSS signals ranging from −20 to 10 dB in terms of signal-to-noise ratios (SNRs). Upon training and verifying the model using BPSK modulation, it is then put to the test with QPSK-modulated signals, and the estimation performance was analyzed through metrics such as loss function, accuracy rate, recall rate, and confusion matrix. The results demonstrate that the 1D CNN-ResNet proposed in this paper is capable of effectively estimating the PN code period of the non-cooperative DSSS signal, exhibiting robust generalization abilities.

With the wide application of direct sequence spread spectrum (DSSS) signals, the comprehensive performance of DSSS communication systems has been continuously improved, making the electronic reconnaissance link in communication countermeasures more difficult. Electronic reconnaissance technology, as the fundamental means of modern electronic warfare, mainly includes signal detection, recognition, and parameter estimation. At present, research on DSSS detection algorithms is mostly based on the correlation characteristics of DSSS signals, and autocorrelation algorithm is the most mature and widely used method in practical engineering. With the continuous development of deep learning, deep-learning-based methods have gradually been introduced to replace traditional algorithms in the field of signal processing. This paper proposes a spread spectrum signal detection method based on convolutional neural network (CNN). Through experimental analysis, the detection performance of the CNN model proposed in this paper on DSSS signals in various situations has been compared and analyzed with traditional autocorrelation detection methods for different signal-to-noise ratios. The experiments verified the estimation performance of the model in this paper under different signal-to-noise ratios, different spreading code lengths, different spreading code types, and different modulation methods and compared it with the autocorrelation detection algorithm. It was found that the detection performance of the model in this paper was higher than that of the autocorrelation detection method, and the overall performance was improved by 4 dB.

Real-time fault detection in power distribution networks has become a popular issue in current power systems. However, the low power and computational capabilities of edge devices often fail to meet the requirements of real-time detection. To overcome these challenges, this paper proposes a lightweight algorithm, named Comprehensive-YOLOv5, for identifying defects in distribution networks. The proposed method focuses on achieving rapid localization and accurate identification of three common defects: insulator without loop, cable detachment from the insulator, and cable detachment from the spacer. Based on the You Only Look Once version 5 (YOLOv5) algorithm, this paper adopts GhostNet to reconstruct the original backbone of YOLOv5; introduces Bidirectional Feature Pyramid Network (BiFPN) structure to replace Path Aggregation Network (PANet) for feature fusion, which enhances the feature fusion ability; and replaces Generalized Intersection over Union GIOU with Focal Extended Intersection over Union (Focal-EIOU) to optimize the loss function, which improves the mean average precision and speed of the algorithm. The effectiveness of the improved Comprehensive-YOLOv5 algorithm is verified through a “morphological experiment”, while an “algorithm comparison experiment” confirms its superiority over other algorithms. Compared with the original YOLOv5, the Comprehensive-YOLOv5 algorithm improves mean average precision (mAP) from 88.3% to 90.1% and increases Frames per second (FPS) from 20 to 52 frames. This improvement significantly reduces false positives and false negatives in defect detection. Consequently, the proposed algorithm enhances detection speed and improves inspection efficiency, providing a viable solution for real-time detection and deployment at the edge of power distribution networks.

Over the past decade, deep learning (DL) has been applied in a large number of optical sensors applications. DL algorithms can improve the accuracy and reduce the noise level in optical sensors. Optical sensors are considered as a promising technology for modern intelligent sensing platforms. These sensors are widely used in process monitoring, quality prediction, pollution, defence, security, and many other applications. However, they suffer major challenges such as the large generated datasets and low processing speeds for these data, including the high cost of these sensors. These challenges can be mitigated by integrating DL systems with optical sensor technologies. This paper presents recent studies integrating DL algorithms with optical sensor applications. This paper also highlights several directions for DL algorithms that promise a considerable impact on use for optical sensor applications. Moreover, this study provides new directions for the future development of related research.

Glacial debris flow is a common natural disaster, and its frequency has been increasing in recent years due to the continuous retreat of glaciers caused by global warming. To reduce the damage caused by glacial debris flows to human and physical properties, glacier susceptibility assessment analysis is needed. Most research efforts consider the effect of existing glacier area and ignore the effect of glacier ablation volume change. In this paper, we consider the impact of glacier ablation volume change to investigate the susceptibility of glacial debris flow. The susceptibility to mudslide was evaluated by taking the glacial mudslide-prone ditch of G318 Linzhi section of Sichuan-Tibet Highway as the research object. First, by using a simple band ratio method with manual correction, we produced a glacial mudslide remote sensing image dataset, and second, we proposed a deep-learning-based approach using a weight-optimized glacial mudslide semantic segmentation model for accurately and automatically mapping the boundaries of complex glacial mudslide-covered remote sensing images. Then, we calculated the ablation volume by the change in glacier elevation and ablation area from 2015 to 2020. Finally, glacial debris flow susceptibility was evaluated based on the entropy weight method and Topsis method with glacial melt volume in different watersheds as the main factor. The research results of this paper show that most of the evaluation indices of the model are above 90%, indicating that the model is reasonable for glacier boundary extraction, and remote sensing images and deep learning techniques can effectively assess the glacial debris flow susceptibility and provide support for future glacial debris flow disaster prevention.

Due to its low stiffness, the boring bar used in deep-hole-boring is prone to violent vibration during the cutting process. It is often inaccurate and inefficient to judge the vibration state of the boring bar through artificial experience. To detect the change of the vibration state of the boring bar over time, guide the adjustment of the processing parameters, and avoid wastage of the workpiece and the loss of equipment, it is particularly important to intelligently monitor the vibration state of the boring bar during processing. In this paper, the boring bar is taken as the research object, and an intelligent monitoring technology of the boring bar’s vibration state based on deep learning is proposed. Based on grouping convolution, channel shuffle, and BiLSTM, a shuffle-BiLSTM NET model is constructed, which is both lightweight and has a high classification accuracy. The boring experiment platform is built, and 192 groups of cutting experiments are carried out. The three-way acceleration and sound pressure signals are collected, and the signals are processed by smoothed pseudo-Wigner–Ville distribution. The original signals are transformed into a 256 × 256 × 3 matrix obtained by a two-dimensional time–frequency spectrum diagram. The matrix is input into the model to recognize the boring bar’s vibration state. The final classification accuracy is 91.2%. A variety of typical deep learning models are introduced for performance comparison, which proves the superiority of the models and methods used in this paper.

This article proposes a CBAM-ASPP-SqueezeNet model based on the attention mechanism and atrous spatial pyramid pooling (CBAM-ASPP) to solve the problem of robot multi-target grasping detection. Firstly, the paper establishes and expends a multi-target grasping dataset, as well as introduces and uses transfer learning to conduct network pre-training on the single-target dataset and slightly modify the model parameters using the multi-target dataset. Secondly, the SqueezeNet model is optimized and improved using the attention mechanism and atrous spatial pyramid pooling module. The paper introduces the attention mechanism network to weight the transmitted feature map in the channel and spatial dimensions. It uses a variety of parallel operations of atrous convolution with different atrous rates to increase the size of the receptive field and preserve features from different ranges. Finally, the CBAM-ASPP-SqueezeNet algorithm is verified using the self-constructed, multi-target capture dataset. When the paper introduces transfer learning, the various indicators converge after training 20 epochs. In the physical grabbing experiment conducted by Kinova and SIASUN Arm, a network grabbing success rate of 93% was achieved.

Lane-level self-localization is essential for autonomous driving. Point cloud maps are typically used for self-localization but are known to be redundant. Deep features produced by neural networks can be used as a map, but their simple utilization could lead to corruption in large environments. This paper proposes a practical map format using deep features. We propose voxelized deep feature maps for self-localization, consisting of deep features defined in small regions. The self-localization algorithm proposed in this paper considers per-voxel residual and reassignment of scan points in each optimization iteration, which could result in accurate results. Our experiments compared point cloud maps, feature maps, and the proposed map from the self-localization accuracy and efficiency perspective. As a result, more accurate and lane-level self-localization was achieved with the proposed voxelized deep feature map, even with a smaller storage requirement compared with the other map formats.

This paper presents a novel deep-learning (DL)-based approach for classifying digitally modulated signals, which involves the use of capsule networks (CAPs) together with the cyclic cumulant (CC) features of the signals. These were blindly estimated using cyclostationary signal processing (CSP) and were then input into the CAP for training and classification. The classification performance and the generalization abilities of the proposed approach were tested using two distinct datasets that contained the same types of digitally modulated signals, but had distinct generation parameters. The results showed that the classification of digitally modulated signals using CAPs and CCs proposed in the paper outperformed alternative approaches for classifying digitally modulated signals that included conventional classifiers that employed CSP-based techniques, as well as alternative DL-based classifiers that used convolutional neural networks (CNNs) or residual networks (RESNETs) with the in-phase/quadrature (I/Q) data used for training and classification.

Fault diagnosis is crucial for repairing aircraft and ensuring their proper functioning. However, with the higher complexity of aircraft, some traditional diagnosis methods that rely on experience are becoming less effective. Therefore, this paper explores the construction and application of an aircraft fault knowledge graph to improve the efficiency of fault diagnosis for maintenance engineers. Firstly, this paper analyzes the knowledge elements required for aircraft fault diagnosis, and defines a schema layer of a fault knowledge graph. Secondly, with deep learning as the main method and heuristic rules as the auxiliary method, fault knowledge is extracted from structured and unstructured fault data, and a fault knowledge graph for a certain type of craft is constructed. Finally, a fault question-answering system based on a fault knowledge graph was developed, which can accurately answer questions from maintenance engineers. The practical implementation of our proposed methodology highlights how knowledge graphs provide an effective means of managing aircraft fault knowledge, ultimately assisting engineers in identifying fault roots accurately and quickly.

Anomaly detection in video surveillance is a highly developed subject that is attracting increased attention from the research community. There is great demand for intelligent systems with the capacity to automatically detect anomalous events in streaming videos. Due to this, a wide variety of approaches have been proposed to build an effective model that would ensure public security. There has been a variety of surveys of anomaly detection, such as of network anomaly detection, financial fraud detection, human behavioral analysis, and many more. Deep learning has been successfully applied to many aspects of computer vision. In particular, the strong growth of generative models means that these are the main techniques used in the proposed methods. This paper aims to provide a comprehensive review of the deep learning-based techniques used in the field of video anomaly detection. Specifically, deep learning-based approaches have been categorized into different methods by their objectives and learning metrics. Additionally, preprocessing and feature engineering techniques are discussed thoroughly for the vision-based domain. This paper also describes the benchmark databases used in training and detecting abnormal human behavior. Finally, the common challenges in video surveillance are discussed, to offer some possible solutions and directions for future research.

Tool wear condition monitoring is an important component of mechanical processing automation, and accurately identifying the wear status of tools can improve processing quality and production efficiency. This paper studied a new deep learning model, to identify the wear status of tools. The force signal was transformed into a two-dimensional image using continuous wavelet transform (CWT), short-time Fourier transform (STFT), and Gramian angular summation field (GASF) methods. The generated images were then fed into the proposed convolutional neural network (CNN) model for further analysis. The calculation results show that the accuracy of tool wear state recognition proposed in this paper was above 90%, which was higher than the accuracy of AlexNet, ResNet, and other models. The accuracy of the images generated using the CWT method and identified with the CNN model was the highest, which is attributed to the fact that the CWT method can extract local features of an image and is less affected by noise. Comparing the precision and recall values of the model, it was verified that the image obtained by the CWT method had the highest accuracy in identifying tool wear state. These results demonstrate the potential advantages of using a force signal transformed into a two-dimensional image for tool wear state recognition and of applying CNN models in this area. They also indicate the wide application prospects of this method in industrial production.

Autonomous vehicles have become a topic of interest in recent times due to the rapid advancement of automobile and computer vision technology. The ability of autonomous vehicles to drive safely and efficiently relies heavily on their ability to accurately recognize traffic signs. This makes traffic sign recognition a critical component of autonomous driving systems. To address this challenge, researchers have been exploring various approaches to traffic sign recognition, including machine learning and deep learning. Despite these efforts, the variability of traffic signs across different geographical regions, complex background scenes, and changes in illumination still poses significant challenges to the development of reliable traffic sign recognition systems. This paper provides a comprehensive overview of the latest advancements in the field of traffic sign recognition, covering various key areas, including preprocessing techniques, feature extraction methods, classification techniques, datasets, and performance evaluation. The paper also delves into the commonly used traffic sign recognition datasets and their associated challenges. Additionally, this paper sheds light on the limitations and future research prospects of traffic sign recognition.

A significant majority of the population in India makes their living through agriculture. Different illnesses that develop due to changing weather patterns and are caused by pathogenic organisms impact the yields of diverse plant species. The present article analyzed some of the existing techniques in terms of data sources, pre-processing techniques, feature extraction techniques, data augmentation techniques, models utilized for detecting and classifying diseases that affect the plant, how the quality of images was enhanced, how overfitting of the model was reduced, and accuracy. The research papers for this study were selected using various keywords from peer-reviewed publications from various databases published between 2010 and 2022. A total of 182 papers were identified and reviewed for their direct relevance to plant disease detection and classification, of which 75 papers were selected for this review after exclusion based on the title, abstract, conclusion, and full text. Researchers will find this work to be a useful resource in recognizing the potential of various existing techniques through data-driven approaches while identifying plant diseases by enhancing system performance and accuracy.