Search

This paper focuses on deep vision sensing-assisted gesture recognition for athletes in dynamic scenes. Although many research attention had been devoted to this field in recent years, most of existing works failed to fully take characteristics of dynamic scenes into consideration. To deal with this challenge, this paper proposes a diffusion convolution neural network-based multiview gesture recognition approach in dynamic scenes. For one thing, the dynamic spatiotemporal slice position selection based on the body mask heatmap is adopted to calculate positions of horizontal and vertical slices. Thus, the dynamic selection of slice positions in two directions can be realized, and then the extraction of bi-directional spatiotemporal slice images can be completed. For another, action sequences through the 3D residual neural network are learned, and the spatiotemporal information among frames are mined through recurrent networks. Through their combination, a multi-view gesture recognition approach for athletes is constructed. In the experiments, two standard datasets UCF101 and HMDB51 are utilized to establish simulation environment. The proposed method can reach the accuracy beyond 95% on the two datasets. Compared with several typical recognition methods, the proposed method shows higher accuracy. [ABSTRACT FROM AUTHOR]

This work addresses the challenge of calibrating multiple solid-state LIDAR systems. The study focuses on three different solid-state LIDAR sensors that implement different hardware designs, leading to distinct scanning patterns for each system. Consequently, detecting corresponding points between the point clouds generated by these LIDAR systems—as required for calibration—is a complex task. To overcome this challenge, this paper proposes a method that involves several steps. First, the measurement data are preprocessed to enhance its quality. Next, features are extracted from the acquired point clouds using the Fast Point Feature Histogram method, which categorizes important characteristics of the data. Finally, the extrinsic parameters are computed using the Fast Global Registration technique. The best set of parameters for the pipeline and the calibration success are evaluated using the normalized root mean square error. In a static real-world indoor scenario, a minimum root mean square error of 7 c m was achieved. Importantly, the paper demonstrates that the presented approach is suitable for online use, indicating its potential for real-time applications. By effectively calibrating the solid-state LIDAR systems and establishing point correspondences, this research contributes to the advancement of multi-LIDAR fusion and facilitates accurate perception and mapping in various fields such as autonomous driving, robotics, and environmental monitoring. [ABSTRACT FROM AUTHOR]

With the rapid advancement of remote-sensing technology, the spectral information obtained from hyperspectral remote-sensing imagery has become increasingly rich, facilitating detailed spectral analysis of Earth's surface objects. However, the abundance of spectral information presents certain challenges for data processing, such as the "curse of dimensionality" leading to the "Hughes phenomenon", "strong correlation" due to high resolution, and "nonlinear characteristics" caused by varying surface reflectances. Consequently, dimensionality reduction of hyperspectral data emerges as a critical task. This paper begins by elucidating the principles and processes of hyperspectral image dimensionality reduction based on manifold theory and learning methods, in light of the nonlinear structures and features present in hyperspectral remote-sensing data, and formulates a dimensionality reduction process based on manifold learning. Subsequently, this study explores the capabilities of feature extraction and low-dimensional embedding for hyperspectral imagery using manifold learning approaches, including principal components analysis (PCA), multidimensional scaling (MDS), and linear discriminant analysis (LDA) for linear methods; and isometric mapping (Isomap), locally linear embedding (LLE), Laplacian eigenmaps (LE), Hessian locally linear embedding (HLLE), local tangent space alignment (LTSA), and maximum variance unfolding (MVU) for nonlinear methods, based on the Indian Pines hyperspectral dataset and Pavia University dataset. Furthermore, the paper investigates the optimal neighborhood computation time and overall algorithm runtime for feature extraction in hyperspectral imagery, varying by the choice of neighborhood k and intrinsic dimensionality d values across different manifold learning methods. Based on the outcomes of feature extraction, the study examines the classification experiments of various manifold learning methods, comparing and analyzing the variations in classification accuracy and Kappa coefficient with different selections of neighborhood k and intrinsic dimensionality d values. Building on this, the impact of selecting different bandwidths t for the Gaussian kernel in the LE method and different Lagrange multipliers λ for the MVU method on classification accuracy, given varying choices of neighborhood k and intrinsic dimensionality d, is explored. Through these experiments, the paper investigates the capability and effectiveness of different manifold learning methods in feature extraction and dimensionality reduction within hyperspectral imagery, as influenced by the selection of neighborhood k and intrinsic dimensionality d values, identifying the optimal neighborhood k and intrinsic dimensionality d value for each method. A comparison of classification accuracies reveals that the LTSA method yields superior classification results compared to other manifold learning approaches. The study demonstrates the advantages of manifold learning methods in processing hyperspectral image data, providing an experimental reference for subsequent research on hyperspectral image dimensionality reduction using manifold learning methods. [ABSTRACT FROM AUTHOR]

Android malware is becoming more common, and its invasion of smart devices has brought immeasurable losses to people's lives. Most existing Android malware detection methods extract Android features from the original application files without considering the high-order hidden information behind them, but these hidden information can reflect malicious behaviors. To solve this problem, this paper proposes Z2F, a detection framework based on multidimensional Android feature extraction and graph neural networks for Android applications. Z2F first extracts seven types of Android features from the original Android application and then embeds them into a heterogeneous graph. On this basis, we design 12 kinds of meta-structures to analyze different semantic spaces of heterogeneous graphs, mine high-order hidden semantic information, and adopt a multi-layer graph attention mechanism to iteratively embed and update information. In this paper, a total of 14429 Android applications were detected and 1039726 Android features were extracted, with a detection accuracy of 99.7%. [ABSTRACT FROM AUTHOR]

With the continuous emergence and development of 3D sensors in recent years, it has become increasingly convenient to collect point cloud data for 3D object detection tasks, such as the field of autonomous driving. But when using these existing methods, there are two problems that cannot be ignored: (1) The bird's eye view (BEV) is a widely used method in 3D objective detection; however, the BEV usually compresses dimensions by combined height, dimension, and channels, which makes the process of feature extraction in feature fusion more difficult. (2) Light detection and ranging (LiDAR) has a much larger effective scanning depth, which causes the sector to become sparse in deep space and the uneven distribution of point cloud data. This results in few features in the distribution of neighboring points around the key points of interest. The following is the solution proposed in this paper: (1) This paper proposes multi-scale feature fusion composed of feature maps at different levels made of Deep Layer Aggregation (DLA) and a feature fusion module for the BEV. (2) A point completion network is used to improve the prediction results by completing the feature points inside the candidate boxes in the second stage, thereby strengthening their position features. Supervised contrastive learning is applied to enhance the segmentation results, improving the discrimination capability between the foreground and background. Experiments show these new additions can achieve improvements of 2.7%, 2.4%, and 2.5%, respectively, on KITTI easy, moderate, and hard tasks. Further ablation experiments show that each addition has promising improvement over the baseline. [ABSTRACT FROM AUTHOR]

Existing biohashing authentication schemes usually store bioinformation in the cloud, whose storage method can face the problems of bioinformation leakage and tampering, and single point of failure. To address the above problems, a ciphertext speech biohashing authentication scheme based on chaotic system and improved public chain is proposed. First, the fusion features are extracted and generated with the inner product of 3D chaotic system to generate the biohashing sequence. Second, the original speech is encrypted and uploaded to the InterPlanetary file system and then, uploaded to the blockchain after the data are encrypted. Finally, the server matches the hash sequence queried on the chain with the to-be-authenticated hash at the user's end using the Hamming distance. The experimental results show that the biohashing algorithm in this paper has better distinguishability, robustness, security and time efficiency. The key space of the speech encryption algorithm is 2 256 , which can well resist the exhaustive attack and ensure the security during speech authentication. In this paper, the combination of on-chain and off-chain storage of speech information can prevent the problem of bioinformation leakage and tampering, as well as avoiding the problem of single point of failure. [ABSTRACT FROM AUTHOR]

Danmakus are user-generated comments that overlay on videos, enabling real-time interactions between viewers and video content. The emotional orientation of danmakus can reflect the attitudes and opinions of viewers on video segments, which can help video platforms optimize video content recommendation and evaluate users' abnormal emotion levels. Aiming at the problems of low transferability of traditional sentiment analysis methods in the danmaku domain, low accuracy of danmaku text segmentation, poor consistency of sentiment annotation, and insufficient semantic feature extraction, this paper proposes a video danmaku sentiment analysis method based on MIBE-RoBERTa-FF-BiLSTM. This paper constructs a "Bilibili Must-Watch List and Top Video Danmaku Sentiment Dataset" by ourselves, covering 10,000 positive and negative sentiment danmaku texts of 18 themes. A new word recognition algorithm based on mutual information (MI) and branch entropy (BE) is used to discover 2610 irregular network popular new words from trigrams to heptagrams in the dataset, forming a domain lexicon. The Maslow's hierarchy of needs theory is applied to guide the consistent sentiment annotation. The domain lexicon is integrated into the feature fusion layer of the RoBERTa-FF-BiLSTM model to fully learn the semantic features of word information, character information, and context information of danmaku texts and perform sentiment classification. Comparative experiments on the dataset show that the model proposed in this paper has the best comprehensive performance among the mainstream models for video danmaku text sentiment classification, with an F1 value of 94.06%, and its accuracy and robustness are also better than other models. The limitations of this paper are that the construction of the domain lexicon still requires manual participation and review, the semantic information of danmaku video content and the positive case preference are ignored. [ABSTRACT FROM AUTHOR]

In industry, the task of defect classification and defect localization is an important part of defect detection system. However, existing studies only focus on one task and it is difficult to ensure the accuracy of both tasks. This paper proposes a defect detection system based on improved Yolo_v4, which greatly improves the detection ability of minor defects. For K_Means algorithm clustering prianchors question with strong subjectivity, the paper proposes the Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to determine the number of Anchors. To solve the problem of low detection rate of small targets caused by insufficient reuse rate of low-level features in CSPDarknet53 feature extraction network, this paper proposes an ECA-DenseNet-BC-121 feature extraction network to improve it. And the Dual Channel Feature Enhancement (DCFE) module is proposed to improve the local information loss and gradient propagation obstruction caused by quad chain convolution in PANet networks to improve the robustness of the model. The experimental results on the fabric surface defect detection datasets show that the mAP of the improved Yolo_v4 is 98.97%, which is 7.67% higher than SSD, 3.75% higher than Faster_RCNN, 10.82% higher than Yolo_v4 tiny, and 5.35% higher than Yolo_v4, and the detection speed reaches 39.4 fps. It can meet the real-time monitoring needs of industrial sites. [ABSTRACT FROM AUTHOR]

For insulated gate bipolar transistor (IGBT) modules using wire bonding as the interconnection method, the main failure mechanism is cracking of the bonded interface. Studying the mechanical properties of the bonded interface is crucial for assessing the reliability of IGBT modules. In this paper, first, shear tests are conducted on the bonded interface to test the bonded interface's strength. Then, finite element–cohesive zone modeling (FE-CZM) is established to describe the mechanical behavior of the bonded interface. A novel machine learning (ML) architecture integrating a convolutional neural network (CNN) and a long short-term memory (LSTM) network is used to identify the shape and parameters of the traction separation law (TSL) of the FE-CZM model accurately and efficiently. The CNN-LSTM architecture not only has excellent feature extraction and sequence-data-processing abilities but can also effectively address the long-term dependency problem. A total of 1800 sets of datasets are obtained based on numerical computations, and the CNN-LSTM architecture is trained with load–displacement (F–δ) curves as input parameters and TSL shapes and parameters as output parameters. The results show that the error rate of the model for TSL shape prediction is only 0.186%. The performance metric's mean absolute percentage error (MAPE) is less than 3.5044% for all the predictions of the TSL parameters. Compared with separate CNN and LSTM architectures, the proposed CNN-LSTM-architecture approach exhibits obvious advantages in recognizing TSL shapes and parameters. A combination of the FE-CZM and ML methods in this paper provides a promising and effective solution for identifying the mechanical parameters of the bonded interfaces of IGBT modules. [ABSTRACT FROM AUTHOR]

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

UAV transmission tower inspection is the use of UAV technology for regular inspection and troubleshooting of towers on transmission lines, which helps to improve the safety and reliability of transmission lines and ensures the stability of the power supply. From the traditional manual tower boarding to the current way of manually selecting target camera shooting points from 3D point clouds to plan the inspection path of the UAV, operational efficiency has drastically improved. However, indoor planning work is still labor-consuming and expensive. In this paper, a deep learning-based point cloud transmission tower segmentation (PCTTS) model combined with the corresponding target point localization algorithm is proposed for automatic segmentation of transmission tower point cloud data and automatically localizing the key inspection component as the target point for UAV inspection. First, we utilize octree sampling with unit ball normalization to simplify the data and ensure translation invariance before putting the data into the model. In the feature extraction stage, we encode the point set information and combine Euclidean distance and cosine similarity features to ensure rotational invariance. On this basis, we adopt multi-scale feature extraction, construct a local coordinate system, and introduce the offset-attention mechanism to enhance model performance further. Then, after the feature propagation module, gradual up-sampling is used to obtain the features of each point to complete the point cloud segmentation. Finally, combining the segmentation results with the target point localization algorithm completes the automatic extraction of UAV inspection target points. The method has been applied to six kinds of transmission tower point cloud data of part segmentation results and three kinds of transmission tower point cloud data of instance segmentation results. The experimental results show that the model achieves mIOU of 94.1% on the self-built part segmentation dataset and 86.9% on the self-built instance segmentation dataset, and the segmentation accuracy outperforms that of the methods for point cloud segmentation, such as PointNet++, DGCNN, Point Transformer, and PointMLP. Meanwhile, the experimental results of UAV inspection target point localization also verify the method's effectiveness in this paper. [ABSTRACT FROM AUTHOR]

Microscopic image analysis technology helps solve the inadvertences of artificial traditional methods in disease, wastewater treatment, and environmental change monitoring analysis. Convolutional neural network (CNN) play an important role in microscopic image analysis. Image segmentation, in which U-Net is increasingly applied in microscopic image segmentation, is a crucial step in detection, tracking, monitoring, feature extraction, modelling, and analysis. This paper comprehensively reviews the development history of U-Net, analyses several research results of various segmentation methods since the emergence of U-Net, and conducts a comprehensive review of related papers. This paper summarised the improved methods of U-Net and then listed the existing significance of image segmentation techniques and their improvements introduced over the years. Finally, focusing on the different improvement strategies of U-Net in different papers, the related work of each application target is reviewed according to detailed technical categories to facilitate future research. Researchers can see the dynamics of the transmission of technological development and keep up with future trends in this interdisciplinary field. [ABSTRACT FROM AUTHOR]

Salient object detection (SOD) aims to identify the most visually striking objects in a scene, simulating the function of the biological visual attention system. The attention mechanism in deep learning is commonly used as an enhancement strategy which enables the neural network to concentrate on the relevant parts when processing input data, effectively improving the model's learning and prediction abilities. Existing saliency object detection methods based on RGB deep learning typically treat all regions equally by using the extracted features, overlooking the fact that different regions have varying contributions to the final predictions. Based on the U2Net algorithm, this paper incorporates the split coordinate channel attention (SCCA) mechanism into the feature extraction stage. SCCA conducts spatial transformation in width and height dimensions to efficiently extract the location information of the target to be detected. While pixel-level semantic segmentation based on annotation has been successful, it assigns the same weight to each pixel which leads to poor performance in detecting the boundary of objects. In this paper, the Canny edge detection loss is incorporated into the loss calculation stage to improve the model's ability to detect object edges. Based on the DUTS and HKU-IS datasets, experiments confirm that the proposed strategies effectively enhance the model's detection performance, resulting in a 0.8% and 0.7% increase in the F1-score of U2Net. This paper also compares the traditional attention modules with the newly proposed attention, and the SCCA attention module achieves a top-three performance in prediction time, mean absolute error (MAE), F1-score, and model size on both experimental datasets. [ABSTRACT FROM AUTHOR]

Since its inception, salient object detection has been widely used in various image processing tasks. However, previous approaches have mainly focused on how to optimize the accuracy of detection, ignoring the computational overhead of the model itself. This makes it difficult to be used in many devices with limited hardware conditions. To address the above problems, this paper proposes a lightweight saliency detection network based on background denoising and feature generation. First, we build a lightweight feature extraction network by improving the MobileNetV3 feature extraction part and with the help of a multidimensional focus module. Subsequently, in order to fully explore the difference between the background and the salient objects. We insert a noise suppression subnet at the end of the encoder. This subnetwork uses the learned background features to reverse fuse the information from the backbone network, effectively decreasing the amount of background noise and encouraging the model to better distinguish subtle differences between salient objects and the background. Finally, in the decoding stage this paper constructs a feature generator that allows the model to efficiently extend the feature map with little increase in the number of parameters. Thereby, it compensates for the lack of expression capability of lightweight model features. The experiment results on six publicly available benchmark datasets prove that our method is more efficient and accurate than existing lightweight saliency detection methods. [ABSTRACT FROM AUTHOR]

Cancelable biometrics serves as an effective countermeasure against various template attacks launched by intruders, safeguarding the biometric system. This paper proposes a cancelable approach with a novel feature extraction technique for iris recognition, known as the hybrid architecture of the convolutional neural network (CNN) and GRU (gated recurrent unit). To provide cancelability to the system, the paper makes use of a random projection technique. The proposed method has the best outcome in terms of accurate identification. The method is validated on two Iris datasets IITD and MMU, which show promising results on the equal error rate (EER) and accuracy. The proposed model provides 0.02 and 0.045 EER for IITD and MMU, respectively, and accuracy 0.98 and 0.933%, for IITD and MMU Iris dataset, respectively, which is very high compared to other methodologies. The proposed hybrid architecture is being used for a cancelable biometric system for the first time based on literature review. The efficiency of the proposed method is high when validated on the datasets. [ABSTRACT FROM AUTHOR]

At present, in end-to-end network combined with deep learning in stereo matching, how to perceive ample image details within an acceptable range of computational cost, so as to improve the final matching accuracy, has become a hot research topic. In this paper, we design a cascaded disparity correction network based on DispNet, which can refine the initial disparity image by combining the process output results in the trunk network. The method proposed in this paper is named ETE–DC–CNN (end to end–disparity correction–convolutional neural network). In the trunk network, the mixed dilated convolution module is employed to extract the features of the original image, and the feature image is constructed into a 3D disparity space volume through depth-separable convolution, after which the initial disparity image can be obtained through 2D encoder–decoder module. In the correction network, the gradient information of the initial feature image is combined to reconstruct the modified matching cost body guided by the initial disparity image and then a smaller encoder–decoder module is trained and processed to make the final result at the sub-pixel level. The algorithm in this paper has been verified in SceneFlow and KITTI 2012/2015 data sets. The method proposed in this article achieves 3.46, 2.12, 1.19, and 0.6 results in the non-occluded area and 4.04, 2.55, 1.58, and 0.7 results in the global area on the Er (> 2 px), Er (> 3 px), Er (> 5 px), and Epe indicators in the KITTI 2012 test set, which are higher than those of the backbone network DispNet. [ABSTRACT FROM AUTHOR]

In view of the problems of high production cost, scarcity and lack of diversity of 3D face datasets, this paper designs an end-to-end self-supervised learning 3D face reconstruction algorithm with a single 2D face image as input, which only uses 2D face datasets to complete model training. First, the improved ResNet module is introduced to preprocess the input face image. The deep residual neural network has strong feature extraction and characterization ability for the image, which can provide rich high-level semantic feature maps for the subsequent subnetwork. Then, add transformer module completely based on self-attention mechanism to the parameter prediction subnetwork, which can make different parameters of the subnetwork focus on self-related feature map information and avoid interference from invalid feature map information, so as to further improve the parameter prediction accuracy of the subnetwork. Next, training, ablation and comparison experiments were conducted on CelebA, BFM and Photoface datasets, and the combined function of pixel loss function and perceptual loss function was selected as the loss function. The experimental results show that: compared with the historical optimal results of the same network structure, the scale-invariant depth error (SIDE) and mean angle deviation (MAD) are improved by 5.9% and 10.8%, respectively, which strongly proves the effectiveness of the algorithm. Finally, in order to verify the actual effect of the 3D face reconstruction algorithm, examples are selected in this paper for reconstruction. The 3D faces generated by the algorithm all have a good sense of reality, which intuitively and effectively proves the advancement of the algorithm. [ABSTRACT FROM AUTHOR]

Recent technological developments have enabled computers to identify and categorize facial expressions to determine a person's emotional state in an image or a video. This process, called "Facial Expression Recognition (FER)", has become one of the most popular research areas in computer vision. In recent times, deep FER systems have primarily concentrated on addressing two significant challenges: the problem of overfitting due to limited training data availability, and the presence of expression-unrelated variations, including illumination, head pose, image resolution, and identity bias. In this paper, a comprehensive survey is provided on deep FER, encompassing algorithms and datasets that offer insights into these intrinsic problems. Initially, this paper presents a detailed timeline showcasing the evolution of methods and datasets in deep facial expression recognition (FER). This timeline illustrates the progression and development of the techniques and data resources used in FER. Then, a comprehensive review of FER methods is introduced, including the basic principles of FER (components such as preprocessing, feature extraction and classification, and methods, etc.) from the pro-deep learning era (traditional methods using handcrafted features, i.e., SVM and HOG, etc.) to the deep learning era. Moreover, a brief introduction is provided related to the benchmark datasets (there are two categories: controlled environments (lab) and uncontrolled environments (in the wild)) used to evaluate different FER methods and a comparison of different FER models. Existing deep neural networks and related training strategies designed for FER, based on static images and dynamic image sequences, are discussed. The remaining challenges and corresponding opportunities in FER and the future directions for designing robust deep FER systems are also pinpointed. [ABSTRACT FROM AUTHOR]

At present, the traditional martial arts teaching method has some problems, such as unstable teaching effect, difficult to quantify, etc. This study aims to use optical technology to improve the effect of martial arts movement teaching. This paper introduces the basic principle and related techniques of human body pose estimation. By analyzing the posture information of the human body, it can accurately capture the movement posture of the students during the martial arts movement, and feed back to the teachers and students in real time. In this paper, a dynamic light acquisition system is designed to capture students' posture information by installing corresponding cameras and sensors in the teaching space. By analyzing and processing the images captured by the camera, the students' posture features can be extracted effectively. Then, according to the characteristics of martial arts movement teaching, the corresponding simulation system is designed. In this system, teachers and students can use virtual reality devices to interact, observe and analyze students' gestures in real time, and make corresponding adjustments and guidance. The conclusion shows that the dynamic light acquisition system based on human body pose estimation has significant advantages in martial arts movement teaching, it can accurately capture students' movement posture and provide real-time feedback and guidance for teachers and students. The application of this technology is expected to improve the effectiveness and quality of martial arts teaching and provide students with a better learning experience. [ABSTRACT FROM AUTHOR]

This research introduces a novel method for emotion recognition using Electroencephalography (EEG) signals, leveraging advancements in emotion computing and EEG signal processing. The proposed method utilizes a semisupervised deep convolutional generative adversarial network (SSDNet) as the central model. The model fully integrates the feature extraction methods of the generative adversarial network (GAN), deep convolutional GAN (DCGAN), spectrally normalized GAN (SSGAN) and the encoder, maximizing the advantages of each method to construct a more accurate emotion classification model. In our study, we introduce a flow-form-consistent merging pattern, which successfully addresses mismatches between the data by fusing the EEG data with the features. Implementing this merging pattern not only enhances the uniformity of the input but also decreases the computational load on the network, resulting in a more efficient model. By conducting experiments on the DEAP and SEED datasets, we evaluate the SSDNet model proposed in this paper in detail. The experimental results show that the accuracy of the proposed algorithm is improved by 6.4% and 8.3% on the DEAP and SEED datasets, respectively, compared to the traditional GAN. This significant improvement in performance validates the effectiveness and feasibility of the SSDNet model. The research contributions of this paper are threefold. First, we propose and implement an SSDNet model that integrates multiple feature extraction methods, providing a more accurate and comprehensive solution for emotion recognition tasks. Second, by introducing a flow-form-consistent merging pattern, we successfully address the problem of interdata mismatches and improve the generalization performance of the model. Finally, we experimentally demonstrate that the method in this paper achieves a significant improvement in accuracy over the traditional GANs on the DEAP and SEED datasets, providing an innovative solution in the field of EEG-based emotion recognition. [ABSTRACT FROM AUTHOR]

Power transformers play a critical role in the performance of power systems. This equipment is costly due to significant copper and iron prices and manufacturing costs. Therefore, maintenance and protection of such equipment is essential. Despite its robust performance, maloperation of differential protection (DP) in transformers may cause operational challenges to power system operators. The differential relay may operate incorrectly after the transformer energization leads to an inrush current (IC) and the relay identifies the event as an internal fault, and consequently issues the trip command. The other case of maloperation includes, but not limited to, a moment when the current transformer saturates due to an external fault. In this paper, a novel approach for DP is proposed, that is based on signal processing methods. In this paper, variational mode decomposition (VMD) and the deep neural network are implemented by using the convolutional neural network (CNN) and bi‐directional long short‐term memory (BiLSTM) models. The VMD decomposes differential current signal (DCS) to intrinsic mode functions with corresponding narrow‐band property frequency spectrums, which provides more detailed information about signal characteristics in different frequency bands. At the next stage, an effective feature for the BiLSTM is extracted by the CNN with the convolutional layers to classify events and proper discrimination. Extensive simulations on a 500 MVA transformer in MATLAB demonstrate the effectiveness of the proposed protection approach to differentiate ICs from internal and external faults with 99.8% accuracy in less than 1/8th of a power cycle. [ABSTRACT FROM AUTHOR]

High-resolution remote sensing image-based vegetation monitoring is a hot topic in remote sensing technology and applications. However, when facing large-scale monitoring across different sensors in broad areas, the current methods suffer from fragmentation and weak generalization capabilities. To address this issue, this paper proposes a multisource high-resolution remote sensing image-based vegetation extraction method that considers the comprehensive perception of multiple features. First, this method utilizes a random forest model to perform feature selection for the vegetation index, selecting an index that enhances the otherness between vegetation and other land features. Based on this, a multifeature synthesis perception convolutional network (MSCIN) is constructed, which enhances the extraction of multiscale feature information, global information interaction, and feature cross-fusion. The MSCIN network simultaneously constructs dual-branch parallel networks for spectral features and vegetation index features, strengthening multiscale feature extraction while reducing the loss of detailed features by simplifying the dense connection module. Furthermore, to facilitate global information interaction between the original spectral information and vegetation index features, a dual-path multihead cross-attention fusion module is designed. This module enhances the differentiation of vegetation from other land features and improves the network's generalization performance, enabling vegetation extraction from multisource high-resolution remote sensing data. To validate the effectiveness of this method, we randomly selected six test areas within Anhui Province and compared the results with three different data sources and other typical methods (NDVI, RFC, OCBDL, and HRNet). The results demonstrate that the MSCIN method proposed in this paper, under the premise of using only GF2 satellite images as samples, exhibits robust accuracy in extraction results across different sensors. It overcomes the rapid degradation of accuracy observed in other methods with various sensors and addresses issues such as internal fragmentation, false positives, and false negatives caused by sample generalization and image diversity. [ABSTRACT FROM AUTHOR]

The emergence of digital media has changed the way people live and learn. In the context of the new era, digital media is gradually integrating into people's life and learning. Digital media contains massive images, and fine-grained image recognition for digital media has become an important topic. The challenge of fine-grained image recognition is that the difference between different categories is small, and the difference between the same categories is sometimes large. This work designs a fine-grained image recognition based on feature enhancement (FIRFE). This extracts as much information as possible from fine-grained images under weak supervision to improve the recognition accuracy. When the existing methods extract image features, the feature extraction other than the most significant local feature is not enough. This deals with local features alone and ignores the relationship between features. First, this paper designs a feature enhancement and suppression module to process image features. Secondly, this paper designs pyramid residual convolution. This uses different scale convolution kernels to capture different levels of features in the scene. Thirdly, this paper uses the softpool method to rationally allocate the information weight in the pooling process. Fourth, this paper uses feature focus module to mine more features. This focuses on obtaining similar information in multiple local features as discriminant features to further improve the recognition. Fifthly, this paper carried out systematic experiments on the designed method. The proposed method achieves 94.3%/95.7% accuracy, 92.9%/94.1% recall, and 91.4%/92.2% F1 score on different datasets. This verified the superiority of this method for fine-grained image recognition of digital media. [ABSTRACT FROM AUTHOR]

This paper introduces a method aiming at enhancing the efficacy of speaker identification systems within challenging acoustic environments characterized by noise and reverberation. The methodology encompasses the utilization of diverse feature extraction techniques, including Mel-Frequency Cepstral Coefficients (MFCCs) and discrete transforms, such as Discrete Cosine Transform (DCT), Discrete Sine Transform (DST), and Discrete Wavelet Transform (DWT). Additionally, an Artificial Neural Network (ANN) serves as the classifier for this method. Reverberation is modeled using varying-length comb filters, and its impact on pitch frequency estimation is explored via the Auto Correlation Function (ACF). This paper also contributes to the field of cancelable speaker identification in both open and reverberation environments. The proposed method depends on comb filtering at the feature level, deliberately distorting MFCCs. This distortion, incorporated within a cancelable framework, serves to obscure speaker identities, rendering the system resilient to potential intruders. Three systems are presented in this work; a reverberation-affected speaker identification system, a system depending on cancelable features through comb filtering, and a novel cancelable speaker identification system within reverbration environments. The findings revealed that, in both scenarios with and without reverberation effects, the DWT-based features exhibited superior performance within the speaker identification system. Conversely, within the cancelable speaker identification system, the DCT-based features represent the top-performing choice. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel method for monocular depth estimation using the hourglass neck module. The proposed method has the following originality. First, feature maps are extracted from Swin Transformer V2 using a masked image modeling (MIM) pretrained model. Since Swin Transformer V2 has a different patch size for each attention stage, it is easier to extract local and global features from images input by the vision transformer (ViT)-based encoder. Second, to maintain the polymorphism and local inductive bias of the feature map extracted from Swin Transformer V2, a feature map is input into the hourglass neck module. Third, deformable attention can be used at the waist of the hourglass neck module to reduce the computation cost and highlight the locality of the feature map. Finally, the feature map traverses the neck and proceeds through a decoder, comprised of a deconvolution layer and an upsampling layer, to generate a depth image. To evaluate the objective reliability of the proposed method in this paper, we used the NYU Depth V2 dataset to compare and evaluate the methods published in other papers. As a result of the experiment, the RMSE value of the novel method for monocular depth estimation using the hourglass neck module proposed in this paper was 0.274, which was lower than those published in other papers. The lower the RMSE value, the better the depth estimation method; therefore, its efficiency compared to other techniques has been proven. [ABSTRACT FROM AUTHOR]

The network structure of digital watermarking algorithm based on deep learning is usually encoder‐noise layer‐decoder. Most of the existing encoders suffer from the problem of insufficient feature extraction, and the introduction of simulated differentiable joint photographic experts group (JPEG) compression in the noise layer cannot ensure the robustness under real JPEG. In this paper, a watermarking algorithm based on multi‐scale auto‐encoder is proposed, which can effectively extract the image feature information by combining with the channel attention mechanism. At the same time, some parameters of decoder and encoder are shared to reduce redundant feature embedding and improve extraction accuracy. This paper also proposes a robust training scheme against JPEG compression, which can guide the model to store the watermark in the low‐frequency region needed for decoding. Experimental results show that the peak signal‐to‐noise ratio (PSNR) of the proposed algorithm is above 48 and the decoding rate is above 99% under JPEG compression with quality factor Q = 50. Moreover, this scheme can effectively promote the combination of noise layer in training. In addition, the proposed algorithm is also robust to other common network noises. [ABSTRACT FROM AUTHOR]

The purpose of this paper is to help people with auditory and speech disabilities to communicate with others and for controlling computers and machines. This paper proposes two different methods for identifying six distinctive hand gestures and sign language for divergent environmental conditions. The first method is based on the hand feature extraction i.e., convexity defects. For that, initially, the hand region is detected by HSV skin color conversion process. Contour and convex hull of hand are extracted from the hand region. Finally, convexity defects are determined to identify the hand gestures. The second method is deep learning based YOLOv3 model that uses DARKNET-53 convolutional neural network (CNN) as its backbone. The model is trained on a large annotated dataset. Experimental results reveal that the deep-leaning method outperforms the hand feature approach and obtain 98.92% and 95.57% accuracy for deep learning and hand feature-based model respectively. [ABSTRACT FROM AUTHOR]

With the advancement of modern medical concepts, the beneficial effects of music on human health have gradually become accepted, and the corresponding music therapy has gradually become a new research direction that has received much attention in recent years. However, folk music has certain peculiarities that lead to the fact that there is no efficient way of selecting repertoire that can be carried out directly throughout the repertoire selection. This paper combines deep learning theory with ethnomusic therapy based on previous research and proposes a deep learning-based approach to ethnomusic therapy song selection. Since the feature extraction process in the traditional sense has insufficient information on each frame, excessive redundancy, inability to process multiple frames of continuous music signals containing relevant music features and weak noise immunity, it increases the computational effort and reduces the efficiency of the system. To address the above shortcomings, this paper introduces deep learning methods into the feature extraction process, combining the feature extraction process of the Deep Auto-encoder (DAE) with the music classification process of Gaussian mixture model, which forms a new DAE-GMM music classification model. Finally, in terms of music therapy selection, this paper compares the music selection method based on co-matrix and physiological signal with the one in this paper. From the theoretical and simulation plots, it can be seen that the method proposed in this paper can achieve both good music classifications from a large number of music and further optimize the process of music therapy song selection from both subjective and objective aspects by considering the therapeutic effect of music on patients. Through this article research results found that the depth of optimization feature vector to construct double the accuracy of the classifier is higher, in addition, compared with the characteristics of the original optimization classification model, using the gaussian mixture model can more accurately classify music, the original landscape "hometown" score of 0.9487, is preferred, insomnia patients mainly ceramic flute style soft tone, without excitant, low depression, have composed of nourishing the heart function. [ABSTRACT FROM AUTHOR]

Monocular depth estimation using a single remote sensing image has emerged as a focal point in both remote sensing and computer vision research, proving crucial in tasks such as 3D reconstruction and target instance segmentation. Monocular depth estimation does not require multiple views as references, leading to significant improvements in both time and efficiency. Due to the complexity, occlusion, and uneven depth distribution of remote sensing images, there are currently few monocular depth estimation methods for remote sensing images. This paper proposes an approach to remote sensing monocular depth estimation that integrates an attention mechanism while considering global and local feature information. Leveraging a single remote sensing image as input, the method outputs end-to-end depth estimation for the corresponding area. In the encoder, the proposed method employs a dense neural network (DenseNet) feature extraction module with efficient channel attention (ECA), enhancing the capture of local information and details in remote sensing images. In the decoder stage, this paper proposes a dense atrous spatial pyramid pooling (DenseASPP) module with channel and spatial attention modules, effectively mitigating information loss and strengthening the relationship between the target's position and the background in the image. Additionally, weighted global guidance plane modules are introduced to fuse comprehensive features from different scales and receptive fields, finally predicting monocular depth for remote sensing images. Extensive experiments on the publicly available WHU-OMVS dataset demonstrate that our method yields better depth results in both qualitative and quantitative metrics. [ABSTRACT FROM AUTHOR]

Space infrared dim target recognition is an important applications of space situational awareness (SSA). Due to the weak observability and lack of geometric texture of the target, it may be unreliable to rely only on grayscale features for recognition. In this paper, an intelligent information decision-level fusion method for target recognition which takes full advantage of the ensemble classifier and Dempster–Shafer (DS) theory is proposed. To deal with the problem that DS produces counterintuitive results when evidence conflicts, a contraction–expansion function is introduced to modify the body of evidence to mitigate conflicts between pieces of evidence. In this method, preprocessing and feature extraction are first performed on the multi-frame dual-band infrared images to obtain the features of the target, which include long-wave radiant intensity, medium–long-wave radiant intensity, temperature, emissivity–area product, micromotion period, and velocity. Then, the radiation intensities are fed to the random convolutional kernel transform (ROCKET) architecture for recognition. For the micromotion period feature, a support vector machine (SVM) classifier is used, and the remaining categories of the features are input into the long short-term memory network (LSTM) for recognition, respectively. The posterior probabilities corresponding to each category, which are the result outputs of each classifier, are constructed using the basic probability assignment (BPA) function of the DS. Finally, the discrimination of the space target category is implemented according to improved DS fusion rules and decision rules. Continuous multi-frame infrared images of six flight scenes are used to evaluate the effectiveness of the proposed method. The experimental results indicate that the recognition accuracy of the proposed method in this paper can reach 93% under the strong noise level (signal-to-noise ratio is 5). Its performance outperforms single-feature recognition and other benchmark algorithms based on DS theory, which demonstrates that the proposed method can effectively enhance the recognition accuracy of space infrared dim targets. [ABSTRACT FROM AUTHOR]

With the rapid development of information transfer technology and the influence of complex electromagnetic environment, the signal components in communication systems are becoming more and more complex, and the spectrum congestion is further aggravated, which seriously affects the performance of communication systems. Single-channel blind source separation technology is an effective solution, but the existing algorithms are deficient in terms of estimation accuracy, time delay and computational complexity. Based on the powerful feature extraction capability and timing signal processing capability of deep learning, this paper proposes an end-to-end time-domain blind separation convolutional network structure, including three parts: encoder, separation module and decoder. The attention mechanism module is also introduced into the encoding layer of the network to enhance the deep feature extraction ability of the network, highlight the influence of important timing features of communication signals, further improve the separation performance and enhance the practicality of the network. The experimental results show that the algorithm has superior performance and can effectively separate the mixed communication signal of various modulation types, and it can also extract own-side communication signals from high-power masked signals. Compared with the existing algorithms, the network in this paper has high separation accuracy and low computational complexity, which breaks the bottleneck of traditional single-channel blind separation algorithms in complex communication electromagnetic environment. [ABSTRACT FROM AUTHOR]

In this paper, a fault feature extraction method based on the combination of computational order tracking (COT) and variational mode decomposition (VMD) is proposed to solve the problem of fault feature extraction in a gearbox under variable speed conditions. First, a speed estimation method based on a forward-backward greedy algorithm and Fourier fitting is proposed to solve the speed estimation problem under variable speed conditions. Then, a feature extraction method based on COT, VMD, and fast kurtometry is proposed. COT is used to calculate the order. After the signal is decomposed by VMD, a fast kurtogram is used to filter the modal components with obvious fault features so as to extract fault features. Finally, combined with the analysis of actual equipment examples, the experimental results show that steady-state filtering can effectively filter the background noise and improve the signal-to-noise ratio. The rotation speed estimated by the forward–backward greedy algorithm and Fourier fitting is very close to the actual speed, which verifies that the method proposed in this paper can effectively solve the problem of fault feature extraction of a gearbox under variable speed conditions. [ABSTRACT FROM AUTHOR]

According to the problem of the current RGB-D feature extraction network used for pose estimation was too large, this paper proposed a lightweight improved XYZNet RGB-D feature extraction network. First, this paper designed a lightweight sub-network BaseNet to replace ResNet18 in XYZNet, which made the network scale significantly reduced and obtained more powerful performance. Then, this paper proposed a re-parameterized multiscale convolutional attention (Rep-MSCA) sub-module based on depth separable convolution, which enhanced the ability of BaseNet to extract contextual information of different scales, and constrained the amount of parameters in the model. Finally, in order to improve the geometric feature extraction ability of PointNet in XYZNet with a small parameter cost, this paper designed a re-parameterized residual multi-layer perceptron (Rep-ResP) module. The floating point operations (FLOPs) and parameters of the improved network are 60.8% and 64.8% lower, the inference speed is 21.2% higher, and the accuracy of the mainstream datasets LineMOD and YCB-Video is 0.5% and 0.6% higher. The proposed model is more suitable for deployment in scenarios where hardware resources are tight. [ABSTRACT FROM AUTHOR]

Automated and accurate detection of power quality (PQ) events is important from the point of view of safety as well as maintaining the reliability of the power transmission and distribution network. However, detection of multiple PQ events in a noisy environment is challenging task. Another important issue is the choice of meaningful features that can directly influence the accuracy of PQ detection. Considering these two aforesaid facts, this paper presents a novel framework for automated classification of PQ signals in a noisy environment employing cross Stockwell Transform (XST). The XST proposed in this paper has better noise suppression capability compared to conventional Stockwell Transform. Here, XST was used to convert 1D PQ signals to 2D time–frequency (T–F) images. To improve the accuracy of PQ detection, an automated feature extraction method employing deep learning is implemented in this work. The noise free T–F images obtained using XST were fed as inputs to several pre-trained convolutional neural networks (CNNs) for deep feature extraction. Transfer learning technique was implemented to reduce the computational cost. The extracted deep features were further undergone selection using one-way analysis of variance test followed by false discovery rate correction. The statistically significant deep features were subsequently fed to three benchmark machine learning classifiers for classification of PQ signals. In addition, tests were also carried out on real-life PQ signals to verify the practicability of the proposed framework. Investigations revealed that the proposed method returned mean accuracy of 99.72% and 96.45% for classification of simulated and real-life PQ signals, respectively. [ABSTRACT FROM AUTHOR]

As a commonly used mode of transportation in people's daily lives, the normal operation of railway transportation is crucial. The track circuit, as a key component of the railway transportation system, is prone to malfunctions due to environmental factors. However, the current method of inspecting track circuit faults still relies on the experience of on-site personnel. In order to improve the efficiency and accuracy of fault diagnosis, we propose to establish an intelligent fault diagnosis system. Considering that the fault data are a one-dimensional time series, this paper presents a fault diagnosis method based on the UNet-LSTM network (ULN). The LSTM network is established on the basis of fault data and used for ZPW-2000A track circuit fault diagnosis. However, the use of a single LSTM network has a high error rate in the common fault diagnosis of track circuits. Therefore, this paper proposes a feature extraction method based on the UNet network. This method is used to extract the features of the original data and then input them into the LSTM network for fault diagnosis. Through experiments with on-site fault data, it has been verified that this method can accurately classify seven common track circuit faults. Finally, the superiority of the method is verified by comparing it with other commonly used fault classification methods. [ABSTRACT FROM AUTHOR]

Based on optical image enhancement, this paper uses convolutional neural network to detect key points of swimming posture. In swimming posture analysis, key point detection can provide information about posture accuracy and evaluation. However, due to the change of illumination conditions, the reduction of image quality and occlusion, the accuracy and stability of key point detection are limited. In order to solve this problem, an optical image enhancement method based on convolutional neural network is proposed in this paper. First, a specific convolutional neural network architecture is used to extract features from input images. Then optical image enhancement technology is used to preprocess the input image to improve the image quality. Next, the trained convolutional neural network model is used to detect key points. Experimental results show that the proposed method has good performance under different illumination conditions and image quality. Therefore, this study provides a new idea and method for the application of optical image enhancement in the detection of key points in swimming posture analysis. [ABSTRACT FROM AUTHOR]

Classification with a few samples of training set has been a long-standing issue in the field of polarimetric synthetic aperture radar (PolSAR) image analysis and processing. In fact, one of the most important challenges of the PolSAR image classification task is the number of labelled samples. In essence, training network by utilizing just few numbers of training samples cannot be led to the reliable result since the neural network is so sensitive to the number of training samples. In addition, in the real PolSAR image classification task, the huge number of training samples is not accessible. On the other hand, classification performance using just small number of labelled samples is not an accurate result. So, in this paper, aiming at the small number of training samples of the PolSAR image classification task, a novel ensemble self-supervised feature-learning (ESSFL) model is designed. The designed ESSFL can automatically extract PolSAR features conducive to PolSAR image classification with a small number of training samples. In addition, it can significantly decrease the dependence of neural network algorithms on large labelled samples of training set. First, to utilize the spatial–polarimetric features of PolSAR data perfectly, the EfficientNet-B0 is presented and utilized as the main section of the deep learning (DL) model to extract DL features of PolSAR data. Then, using an optimization function that constrains the cross-correlation matrix of various distortions of each sample to the identity matrix, the designed deep learning model can obtain the effective features of homogeneous samples gathering together and heterogeneous samples separating from each other in a self-supervised manner. Moreover, two ensemble learning (EL) models, feature-level and view-level ensemble, are proposed to increase the feature extraction capability and classification result by jointly using spatial features at different scales and polarimetric information at different bands. Finally, the stack of the obtained features and the main polarimetric information of PolSAR data can be input into classifier for the classification of PolSAR data. In this paper we use two different classifiers, the first one is the Random Forest (RF) classifier and the second one is the Support Vector Machine (SVM) classifier. The advantages of SVM include that it can be used to avoid the difficulties of using linear functions in the high-dimensional feature space, and the optimization problem is transformed into dual convex quadratic programs that are so usable for an actual image classification task. In addition, among all the available classification methods, random forests provide one of the highest accuracies. The random forest technique can also handle big data with numerous variables running into thousands. In fact, it can automatically balance data sets when a class is more infrequent than other classes in the data that is so important for a real PolSAR image classification task. Experimental results on three well-known PolSAR data sets illustrate that the designed ESSFL can extract more discriminant features using the designed deep learning model. In the end, the experiments prove that the designed ESSFL has a significant classification performance compared with different deep learning models in the case of small number of training samples, and also it can achieve a better result in the case of large number of training samples by comparison with the most of the deep learning models. [ABSTRACT FROM AUTHOR]

The main purpose of this paper is to provide information on how to create a convolutional neural network (CNN) for extracting features from EEG signals. Our task was to understand the primary aspects of creating and fine-tuning CNNs for various application scenarios. We considered the characteristics of EEG signals, coupled with an exploration of various signal processing and data preparation techniques. These techniques include noise reduction, filtering, encoding, decoding, and dimension reduction, among others. In addition, we conduct an in-depth analysis of well-known CNN architectures, categorizing them into four distinct groups: standard implementation, recurrent convolutional, decoder architecture, and combined architecture. This paper further offers a comprehensive evaluation of these architectures, covering accuracy metrics, hyperparameters, and an appendix that contains a table outlining the parameters of commonly used CNN architectures for feature extraction from EEG signals. [ABSTRACT FROM AUTHOR]

In the wake of the development and advancement of signal processing technology for communication radiation source individual, Signal fingerprint feature extraction and analysis technology for communication radiation source individual has broad application prospects in many fields. To effectively extract the individual characteristics of different modulated signals under low SNR environment, and recognize the subtle features for communication radiation source individuals has been a hot spot. Aiming at the problem of signal feature extraction and classifier design under low SNR environment, in the paper, a multifractal dimension and improved gray relation theory based classifier design algorithm is proposed. Firstly, the multifractal dimension feature extraction of nine modulated communication signals is realized. Then multifractal dimension features of these modulated signals under different SNR are compared. An improved gray relation algorithm is used to recognize the extracted subtle characteristics. Meanwhile, FSK signal is used to simulate radio subtle features by adding different distribution of noise. Subtle feature extraction by means of multifractal dimension algorithm and pattern recognition by means of improved gray relation algorithm are used to test the effectiveness of the proposed method for identification of modulated signals and radio subtle features. The simulation results show that the recognition success rate of nine different communication modulated signals can reach 93% even under the SNR of 2 dB, and the recognition success rate of the subtle features of distributed noise can reach 100%. The proposed method provides an effective theoretical basis for identifying of radio modulated signals and communication radiation source individual subtle features. [ABSTRACT FROM AUTHOR]

X-ray security inspection for detecting prohibited items is widely used to maintain social order and ensure the safety of people's lives and property. Due to the large number of parameters and high computational complexity, most current object detection models are challenging to deploy on portable mobile security inspection devices. Therefore, this paper proposes an improved YOLOv7-tiny model for application in prohibited item detection tasks. Firstly, the feature extraction backbone network is replaced with the lightweight GhostNet network to reduce computational complexity and improve detection speed. Secondly, the FPN in the Neck is replaced with BiFPN, further reducing computational complexity and memory access through skip connections. Finally, a lightweight CA attention mechanism is embedded between the Backbone and Neck layers, and the Focal-EIoU Loss function is employed to enhance the detection capability for small-sized items. Experimental results on the SIXray public dataset show a 14.8% reduction in model parameters, a 19.7% reduction in computational complexity, and a 15.9% reduction in volume after the improvements. The detection speed increases from 82.4 to 90.2, and the detection accuracy for prohibited items reaches 90.3%. The experimental results demonstrate that the improved model achieves overall lightweighting while maintaining a high detection rate and improving detection speed. [ABSTRACT FROM AUTHOR]

Plant diseases and pests represent significant threats to agricultural yields, underscoring the need for precise and prompt identification of pathogens and pests. In this paper, we propose a lightweight convolutional neural network model called Gated Asymmetrical GhostNet (GA-GhostNet), specifically designed for the automatic identification of plant diseases and pests. The model incorporates a Gated Multi-scale Coordinate Attention (GM-CA) module to filter out noise and irrelevant information in the image, while capturing location information of diseases and pests at different scales. The model also utilizes an improved feature extraction module called Asymmetrical Ghost (AG) module to enhance robustness to image flipping, as well as improve feature extraction capabilities. Additionally, the CutMix data augmentation method is employed to improve generalization ability. In extensive experimental evaluations on the IP102 pest dataset, GA-GhostNet achieved impressive results with 63.73% MRec, 67.37% MPre, 65.12% MF1, and 71.90% Acc. Moreover, through transfer learning on the Jute pest, Embrapa disease, and Apple disease datasets, the model outperforms numerous lightweight models, attaining accuracies of 99.89%, 96.97%, and 95.17%, respectively, while having only 3.73 million parameters. GA-GhostNet demonstrates high accuracy and efficiency in identifying plant diseases and pests. [ABSTRACT FROM AUTHOR]

In recent years, with the development of deep learning, SAR image-based ship target-detection technology has become a current research hotspot. SAR images are characterized by complex backgrounds, significant speckle noise interference, and poor interpretability. To address this challenge, this study proposes a new Gaussian circle radius determination strategy and a multi-level feature fusion codec-based ship detection network (MFFC-SDN) for SAR images. Differing from CenterNet's corner-oriented Gaussian circle strategy, the Gaussian circle strategy proposed in this paper takes the centre point coordinates of the real target as the origin to define a circular region. This approach allows for more precise supervision of the training model. Meanwhile, we propose MFFC-SDN to solve the problem of the poor performance of CenterNet in SAR ship detection. MFFC-SDN is an anchor-free, single-stage target-detection method with a low computation cost. In MFFC-SDN, we introduce a Multilevel Feature Recapture Codec Module (MFR-CM) to enhance the feature extraction capability of the network by using the features of the encoding stage twice because of the easy loss of image features in the lengthy codec process of the CenterNet network backbone. A Residual Attention Pyramid Feature Fusion Module (RA-PFFM) is designed to enhance the feature map and obtain multi-scale features. Experimental results on the SSDD and SAR-ship-dataset show that MFFC-SDN significantly enhances SAR image target-detection performance, effectively addressing detection challenges related to large target size variations and complex environmental conditions. Compared with existing algorithms, MFFC-SDN achieves the highest $A{P^{50}}$ A P 50 . [ABSTRACT FROM AUTHOR]

Accurate cloud and shadow detection is a crucial prerequisite for optical remote sensing image analysis and application. Multi-temporal-based cloud and shadow detection methods are a preferable choice to detect clouds in complex scenes (e.g., thin clouds, broken clouds and clouds with interference from artificial surfaces with high reflectivity). However, such methods commonly require cloud-free reference images, and this may be difficult to achieve in time-series data since clouds are often prevalent and of varying spatial distribution in optical remote sensing images. Furthermore, current multi-temporal-based methods have limited feature extraction capability and rely heavily on prior assumptions. To address these issues, this paper proposes a Siamese network (Siamnet) for cloud and shadow detection based on Time-Series cloudy Images, namely TSI-Siamnet, which consists of two steps: 1) low-rank and sparse component decomposition of time-series cloudy images is conducted to construct a composite reference image to cope with dynamic changes in the cloud distribution in time-series images; 2) an extended Siamnet with optimal difference calculation module (DM) and multi-scale difference features fusion module (MDFM) is constructed to extract reliable disparity features and alleviate semantic information feature dilution during the decoder part. TSI-Siamnet was tested extensively on seven land cover types in the well-known Landsat 8 Biome dataset. Compared to six state-of-the-art methods (including four deep learning-based methods and two classical non-deep learning-based methods), TSI-Siamnet produced the best performance with an overall accuracy of 95.05% and MIoU of 84.37%. In three more challenging experiments, TSI-Siamnet showed enhanced detection of thin and broken clouds and greater anti-interference to highly reflective surfaces. TSI-Siamnet provides a novel strategy to explore comprehensively the valid information in time-series cloudy images and integrate the extracted spectral-spatial–temporal features for reliable cloud and shadow detection. [ABSTRACT FROM AUTHOR]

YOLOv7 is one of the most effective algorithms for one‐stage detectors. However, when it is applied to pedestrian detection tasks in the industrial scene, it is still challenging for complex environments and multi‐scale changes of pedestrians. This paper proposes a new pedestrian detector for the industrial scene based on improved YOLOv7‐tiny and named as GP‐YOLO. First, the neck of YOLOv7‐tiny is replaced by RepGFPN structure, make full use of multi‐scale features to enhance the detection accuracy of objects with large‐scale changes. Second, a new gnconv branch is added to the feature fusion module, and the high‐order spatial interaction capability is introduced to further enhance the target detection accuracy. Finally, a lightweight method based on PModule is proposed, on this basis, a PConv bottleneck is designed to reduce the FLOPs and enhance the feature extraction. Experiments on a self‐made Industrial Pedestrian Data set show that before lightweight, the proposed algorithm achieves a 3.2% improvement in mAP@0.5:0.95 and a 3.7% improvement in Recall compared to the baseline YOLOv7‐tiny. After lightweight GP‐YOLO, compared to non‐lightweight, parameters and FLOPs are decreased by 26% and 23%, respectively, the mAP@0.5:0.95 is decreased by only 1.1% and the Recall is decreased by only 1.3%, which remains at a high level. Compared with baseline YOLOv7‐tiny, the lightweight GP‐YOLO has similar parameters and FLOPs, but the mAP@0.5:0.95 is increased by 2.1%, and the Recall is increased by 2.4%. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Pedestrian behavior exhibits high levels of dynamism, and pedestrian trajectories are influenced not only by the pedestrians themselves, but also by interactions with surrounding objects. Efficiently understanding pedestrian motion behavior and modeling its interactions play a crucial role in fields like autonomous driving. Addressing issues related to dynamic feature extraction and interaction modeling in pedestrian trajectory prediction tasks, this paper introduces the conditional pedestrian trajectory prediction network (CSGAT-Net) based on semantic segmentation maps and spatiotemporal graph attention. CSGAT-Net models the physical environment and pedestrian behavior information in the scene as a semantic map, and it leverages graph attention networks to extract pedestrian interaction features. Finally, it predicts pedestrian future trajectories using a variational autoencoder. Comparative experiments conducted on publicly available datasets, ETH and UCY, show that our model exhibits favorable objective evaluation metrics and subjective prediction performance. Particularly, in terms of ADE and FDE metrics, CSGAT-Net outperforms current state-of-the-art methods, indicating that our model can reasonably and accurately predict pedestrian trajectories in different scenarios. [ABSTRACT FROM AUTHOR]

An energy-saving control strategy for wheel loaders is proposed in this paper to address the issue of high energy consumption during their operation. The strategy is based on the intelligent identification of working stages, allowing for staged power matching and resulting in reduced energy consumption. Each work stage of the loader is identified by matching it to the main pump pressure waveform and actuator pilot pressure waveform. Using a sliding time window method, pressure waveforms from each working stage are subjected to feature extraction. A bidirectional long short-term memory neural network (BILSTM) algorithm is then used to establish an intelligent recognition model. Based on work stage identification, an energy-saving control strategy based on power matching is proposed for the shoveling stage of the loader, and the Grey Wolf optimization (GWO)-PID algorithm is utilized for control parameter tuning. Finally, the effectiveness of the energy-saving control strategy based on work stage identification is verified through experiments. The research results indicate that the BILSTM recognition model outperforms other models with a recognition accuracy of 96.1%. The optimal time window width is 0.6 s, and the proposed energy-saving control strategy achieves a fuel-saving rate of 6.81%. This method provides feasibility for reducing energy consumption in construction machinery and achieving energy-saving and carbon-reduction goals. [ABSTRACT FROM AUTHOR]

Lung cancer is an uncontrolled growth of tissue causing a lump in the human lung. If lung cancer can be detected early, it can increase the survival rate. Therefore, a multi-classification approach of lung nodule detection with high computational effectiveness is required. In this paper, a multi-classification approach of lung nodule detection and classification is proposed using artificial intelligence on computed tomography (CT) scan images. Different preprocessing steps are applied for resizing, smoothing, and enhancement of the CT images. Then, two different approaches for feature extraction using VGG16 transfer learning and morphological segmentation are proposed. Morphological segmentation and feature extraction are applied for the segmentation of the region of interest and to extract the distinct features. Finally, the proposed deep learning architecture and seven different machine learning algorithms are applied on the preprocessed data and the extracted features for the classification of lung nodules into three classes: malignant, benign, and normal. It is observed that the stacked ensemble model of deep learning convolutional neural network (CNN) and VGG16 transfer learning models (CNN+VGG16) can achieve 99.55% accuracy using preprocessed data. It is also observed that all the ML algorithms perform with reasonably high accuracy using the low-dimensional morphological features. It is observed from the fivefold cross-validation results that logistic regression performs with 99.36% accuracy in 23.71 s time using the preprocessed data. Whereas, using the morphological features, k-nearest neighbor, and the support vector machine perform with the highest accuracy of 99.76% with very reduced computational time of 0.017 and 0.008 s, respectively. [ABSTRACT FROM AUTHOR]