Search

Airborne surface and underwater human pose recognition are crucial for various safety and surveillance applications, including the detection of individuals in distress or drowning situations. However, airborne optical cameras struggle to achieve simultaneous imaging of the surface and underwater because of limitations imposed by visible-light wavelengths. To address this problem, this study proposes the use of light detection and ranging (LiDAR) to simultaneously detect humans on the surface and underwater, whereby human poses are recognized using a neural network designed for irregular data. First, a temporal point-cloud dataset was constructed for surface and underwater human pose recognition to enhance the recognition of comparable movements. Subsequently, radius outlier removal (ROR) and statistical outlier removal (SOR) were employed to alleviate the impact of noise and outliers in the constructed dataset. Finally, different combinations of secondary sampling methods and sample sizes were tested to improve recognition accuracy using PointNet++. The experimental results show that the highest recognition accuracy reached 97.5012%, demonstrating the effectiveness of the proposed human pose detection and recognition method., (© 2024. The Author(s).)

Pine wilt disease (PWD) is one of the most concerning diseases in forestry and poses a considerable threat to forests. Since the deep learning approach can interpret the raw images acquired by UAVs, it provides an effective means for forest health detection. However, the fact that only PWD can be detected but not the degree of infection can be evaluated hinders forest management, so it is necessary to establish an effective method to accurately detect PWD and extract regions infected by PWD. Therefore, a Mask R-CNN-based PWD detection and extraction algorithm is proposed in this paper. Firstly, the extraction of image features is improved by using the advanced ConvNeXt network. Then, it is proposed to change the original multi-scale structure to PA-FPN and normalize it by using GN and WS methods, which effectively enhances the data exchange between the bottom and top layers under low Batch-size training. Finally, a branch is added to the Mask module to improve the ability to extract objects using fusion. In addition, a PWD region extraction module is proposed in this paper for evaluating the damage caused by PWD. The experimental results show that the improved method proposed in this paper can achieve 91.9% recognition precision, 90.2% mapping precision, and 89.3% recognition rate of the affected regions on the PWD dataset. It can effectively identify the distribution of diseased pine trees and calculate the damage proportion in a relatively accurate way to facilitate the management of forests. [ABSTRACT FROM AUTHOR]

Dedicated equipment, which is widely used in many different types of vehicles, is the core system that determines the combat capability of special vehicles. Therefore, assuring the normal operation of dedicated equipment is crucial. With the increase in battlefield complexity, the demand for equipment functions is increasing, and the complexity of dedicated equipment is also increasing. To solve the problem of fault diagnosis of dedicated equipment, a fault diagnosis algorithm based on CNN-LSTM was proposed in this paper. CNN and LSTM are used in the model adopted by the algorithm to extract spatial and temporal features from the data. CBAM is used to enhance the model's accuracy in identifying faults for dedicated equipment. Data on dedicated equipment faults were obtained from a hardware-in-loop simulation platform to verify the model. It is demonstrated that the proposed fault diagnosis algorithm has high recognition ability for dedicated equipment by comparing it to other neural network models. [ABSTRACT FROM AUTHOR]

Grain yield prediction affects policy making in various aspects such as agricultural production planning, food security assurance, and adjustment of foreign trade. Accurately predicting grain yield is of great significance in ensuring global food security. This paper is based on the MODIS remote sensing image data products from 2010 to 2020, and adds band information such as vegetation index and temperature to form composite remote sensing data as a dataset. Aiming at the lack of models for large-scale forecasting and the need for human intervention in traditional models, this paper proposes a grain production estimation model based on deep learning. First, image cropping and yield mapping techniques are used to process the data to generate training samples. Then the channel and spatial attention mechanism (convolutional block attention module, CBAM) is added to extract spatial information in different remote sensing bands to improve the efficiency of the model. Long short-term memory (LSTM) neural networks are added to obtain feature information in the time dimension. Finally, a national-scale grain yield prediction model is constructed. After the study, it was found that the LSTM model using a combination of multi-source satellite images and an attention mechanism can effectively predict grain yield in China. Furthermore, the proposed model was tested on data from 2018 to 2020 showing an average R 2 of 0.940 and an average RMSE of 80,020 tons, indicating that it can predict Chinese grain yield better. The model proposed in this paper extracts grain yield information directly from the composite remote sensing data, and solves the problem of small-scale research and imprecise yield prediction in an end-to-end manner. [ABSTRACT FROM AUTHOR]

The health of the trees in the forest affects the ecological environment, so timely detection of Standing Dead Trees (SDTs) plays an important role in forest management. However, due to the large spatial scope of forests, it is difficult to find SDTs through conventional approaches such as field inventories. In recent years, the development of deep learning and Unmanned Aerial Vehicle (UAV) has provided technical support for low-cost real-time monitoring of SDTs, but the inability to fully utilize global features and the difficulty of small-scale SDTs detection have brought challenges to the detection of SDTs in visible light images. Therefore, this paper proposes a multi-scale attention mechanism detection method for identifying SDTs in UAV RGB images. This method takes Faster-RCNN as the basic framework and uses Swin-Transformer as the backbone network for feature extraction, which can effectively obtain global information. Then, features of different scales are extracted through the feature pyramid structure and feature balance enhancement module. Finally, dynamic training is used to improve the quality of the model. The experimental results show that the algorithm proposed in this paper can effectively identify the SDTs in the visible light image of the UAV with an accuracy of 95.9%. This method of SDTs identification can not only improve the efficiency of SDTs exploration, but also help relevant departments to explore other forest species in the future. [ABSTRACT FROM AUTHOR]

Recent research advances in salient object detection (SOD) could largely be attributed to ever-stronger multi-scale feature representation empowered by the deep learning technologies. The existing SOD deep models extract multi-scale features via the off-the-shelf encoders and combine them smartly via various delicate decoders. However, the kernel sizes in this commonly-used thread are usually “fixed”. In our new experiments, we have observed that kernels of small size are preferable in scenarios containing tiny salient objects. In contrast, large kernel sizes could perform better for images with large salient objects. Inspired by this observation, we advocate the “dynamic” scale routing (as a brand-new idea) in this paper. It will result in a generic plug-in that could directly fit the existing feature backbone. This paper’s key technical innovations are two-fold. First, instead of using the vanilla convolution with fixed kernel sizes for the encoder design, we propose the dynamic pyramid convolution (DPConv), which dynamically selects the best-suited kernel sizes w.r.t. the given input. Second, we provide a self-adaptive bidirectional decoder design to accommodate the DPConv-based encoder best. The most significant highlight is its capability of routing between feature scales and their dynamic collection, making the inference process scale-aware. As a result, this paper continues to enhance the current SOTA performance. Both the code and dataset are publicly available at https://github.com/wuzhenyubuaa/DPNet. [ABSTRACT FROM AUTHOR]

We investigate privacy-preserving, video-based action recognition in deep learning, a problem with growing importance in smart camera applications. A novel adversarial training framework is formulated to learn an anonymization transform for input videos such that the trade-off between target utility task performance and the associated privacy budgets is explicitly optimized on the anonymized videos. Notably, the privacy budget, often defined and measured in task-driven contexts, cannot be reliably indicated using any single model performance because strong protection of privacy should sustain against any malicious model that tries to steal private information. To tackle this problem, we propose two new optimization strategies of model restarting and model ensemble to achieve stronger universal privacy protection against any attacker models. Extensive experiments have been carried out and analyzed. On the other hand, given few public datasets available with both utility and privacy labels, the data-driven (supervised) learning cannot exert its full power on this task. We first discuss an innovative heuristic of cross-dataset training and evaluation, enabling the use of multiple single-task datasets (one with target task labels and the other with privacy labels) in our problem. To further address this dataset challenge, we have constructed a new dataset, termed PA-HMDB51, with both target task labels (action) and selected privacy attributes (skin color, face, gender, nudity, and relationship) annotated on a per-frame basis. This first-of-its-kind video dataset and evaluation protocol can greatly facilitate visual privacy research and open up other opportunities. Our codes, models, and the PA-HMDB51 dataset are available at: https://github.com/VITA-Group/PA-HMDB51 [ABSTRACT FROM AUTHOR]

Bearing fault diagnosis suffers from class imbalances and distributional discrepancies of fault data under different working conditions. The class imbalance of the fault class increases the difficulty of learning the classification boundary of the diagnostic model for the minority class. Furthermore, the diversities of feature distributions decrease the diagnostic model's generalizability under various working conditions. Thus, in this paper, a deep adversarial transfer learning model for imbalanced bearing fault diagnosis (deep Imba-DA) is proposed to overcome these problems. In the proposed method, a cost-sensitive deep classifier is used to solve the class imbalance problem, and the domain adversarial subnet with the intraclass maximum mean discrepancy (MMD) is used to minimize the marginal and conditional distributional discrepancy between the source (data under one working condition) and target domain (data under another working condition) simultaneously. The performance of deep Imba-DA is evaluated and analyzed on the Case Western Reserve University (CWRU) and Paderborn datasets. The results show that deep Imba-DA outperforms other baseline methods on bearing diagnostic tasks. • An imbalanced bearing fault diagnostic model under variant conditions is proposed. • A joint transfer network with marginal and conditional distribution is developed. • Pseudo label strategy guarantees a good accuracy and fast convergence. [ABSTRACT FROM AUTHOR]

• We utilize parallel sub networks to automatically reveal saliency clues at different spatial levels. • We propose an end-to-end salient object detection model that learns diversity saliency clues in an iterative manner. • We also provide a novel selective fusion strategy to fuse multi-model saliency clues for a high-performance salient object detection. Fully convolutional networks have shown outstanding performance in the salient object detection (SOD) field. The state-of-the-art (SOTA) methods have a tendency to become deeper and more complex, which easily homogenize their learned deep features, resulting in a clear performance bottleneck. In sharp contrast to the conventional "deeper" schemes, this paper proposes a "wider" network architecture which consists of parallel sub-networks with totally different network architectures. In this way, those deep features obtained via these two sub-networks will exhibit large diversity, which will have large potential to be able to complement with each other. However, a large diversity may easily lead to the feature conflictions, thus we use the dense short-connections to enable a recursively interaction between the parallel sub-networks, pursuing an optimal complementary status between multi-model deep features. Finally, all these complementary multi-model deep features will be selectively fused to make high-performance salient object detections. Extensive experiments on several famous benchmarks clearly demonstrate the superior performance, good generalization, and powerful learning ability of the proposed wider framework. [ABSTRACT FROM AUTHOR]

To solve the problems of poor exploration ability and convergence speed of traditional deep reinforcement learning in the navigation task of the patrol robot under indoor specified routes, an improved deep reinforcement learning algorithm based on Pan/Tilt/Zoom(PTZ) image information was proposed in this paper. The obtained symmetric image information and target position information are taken as the input of the network, the speed of the robot is taken as the output of the next action, and the circular route with boundary is taken as the test. The improved reward and punishment function is designed to improve the convergence speed of the algorithm and optimize the path so that the robot can plan a safer path while avoiding obstacles first. Compared with Deep Q Network(DQN) algorithm, the convergence speed after improvement is shortened by about 40%, and the loss function is more stable. [ABSTRACT FROM AUTHOR]

Ultrasonic testing (UT) is commonly used to inspect the geometric shape of internal damage in composite materials and the test results need to be interpreted by trained experts. In this work, an automatic signal classification method based on deep learning is proposed for depth estimation of the detects introduced by low-velocity impact (LVI) in carbon fiber reinforced plastics (CFRPs). Three kinds of neural networks, LSTM, CNN, and CNN-LSTM are used to analyze the attributes with different depths. Then, trained models are applied to identify the depth information of impact damage. The results show that the CNN-LSTM model is a more accurate in-depth classification for LVI defects in CFRP based on A-scan signals than the other two structures. [ABSTRACT FROM AUTHOR]

• A modified-YOLOv5s was proposed to detect defects of wire braided hoses. • K-means++ clustering algorithm is adopted to obtain more suitable anchor boxes. • Focal loss function was adopted to balance between negative and positive samples. • New tailored prediction heads and efficient channel attention (ECA) mechanism were used to improve the detection performance. High-pressure steel wire braided hoses have periodic braided steel wire structure and its productivity is too huge to refrain from making mistakes occasionally such as wire missed wire stacked and wire loosened. The major sticking points in detection are ultra-small, structural similarity and high reflectance of the steel wire braided hoses. Therefore, this paper proposed an automated method based on YOLOv5 to replace traditional naked eyes detection. The anchor boxes were optimized by K-means++ algorithm for defects dataset. Focal loss was used to mitigate the impact of samples imbalance. The predict heads were modified to increase the detection accuracy of ultra-small targets with guaranteeing speed. The effect of efficient channel attention mechanism (ECA) in different positions was analyzed to achieve a better network and increase detection performance. The experimental results showed that the accuracy and recognition efficiency of the proposed modified-YOLOv5s model could reach 92.2% and 23 frames per second. [ABSTRACT FROM AUTHOR]

• A temperature component is given by theoretical inference and parameter optimization. • Separate modeling technique is proposed based on ICEEMDAN and clustering analysis. • The ET-SMT-DELM model is established by a multi-output deep extreme learning machine. • The proposed model has high accuracy, credibility and interpretability. The anatomy of service status through health monitoring models is essential to long-term structural safety. Since deformation is the intuitive representation of the dam's operating condition, it is crucial to investigate the deformation monitoring model with high accuracy and strong interpretability for the safety management of high arch dams. A novel deformation monitoring model is proposed by incorporating impulse response-based equivalent temperature (ET) and machine learning-aided separate modeling technique (SMT). This methodology has three main sources of novelty. First, the impulse response-based equivalent temperature and corresponding temperature component are derived from heat transfer and temperature convolution theories. Then, the model parameters are identified by an improved firefly algorithm. Second, the components in the monitored displacement are progressively stripped out by clustering analysis, separation under equal water level conditions, and a robust signal decomposition technique. Third, the separated deformation components and environmental factors are modeled by the multi-output deep extreme learning machine (DELM) with autoencoder, and the monitoring model ET-SMT-DELM is thus established. The world's highest arch dam is selected to illustrate the proposed model, and the prediction accuracy, early warning performance, component shares, and impulse response mechanism are comprehensively investigated by comparing with several typical baseline models. The results show that the overfitting of the proposed model is reduced, and the prediction and early warning performance is significantly improved. The resulting temperature impulse response function and component shares imply that the interpretability of the model is also enhanced. [ABSTRACT FROM AUTHOR]

Accurate 3D representations of multiphase layered composite fabric, including distinguish and label the pores, matrices, warp and weft yarns, can assist in understanding and ultimately improving composite material design performance. In this work, a new data enhancement algorithm is proposed to generate realistic-looking artificial learning datasets for expanding the datasets. Then, we demonstrate how supervised, the Swin Transformer approach can help to realize accurate segmentation of low-resolution and poor-contrast fabric datasets by populating databases with real and artificial learning datasets. Most of the pixels were correctly identified and the boundaries of the yarn could be identified clearly, except for some spots and a few merged boundaries. The work presented here provides an effective tool that can be widely used in semantic segmentation for various composite materials (woven composites, fiber-reinforced composites, short-fiber materials, etc.), which enables higher-precision reconstruction of fiber-composites and significant reduction in dataset processing time, and provides a route to generating artificial learning dataset. [ABSTRACT FROM AUTHOR]