WWWei Wang

Wei Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Wei Wang has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A point cloud registration network for highly rotationally symmetric components with holes

    by Xiaoxiao Wei, Jinsong Du, Mingrui Fu, Wei Wang, Xingqiang Li, Liang Gao, Shuai Chen

  • A Domain Generalization Framework for Online Aero-Engine Rub-Impact Detection Using Adversarial Learning and Multi-Source Label Aggregation

    by Haotian Peng, Jie Gao, Yafeng Shen, Wei Wang, Chunyan Hu, Jinsong Du

    Online monitoring of rub-impact faults is essential for aero-engine safety, yet existing methods struggle with asynchronous multi-modal data, diverse sensor configurations, and the scarcity of reliable fault annotations. To address these challenges, this paper proposes RubNet, a domain generalization framework that first employs a physics-guided temporal alignment strategy to precisely align high-frequency vibration signals with low-frequency state parameters based on spectral centroid correlations. A multi-scale channel attention convolutional neural network is subsequently utilized to extract robust features, leveraging adversarial domain generalization to learn representations that remain invariant to shifts in sensor position and orientation. Furthermore, to mitigate label noise during temporal fine-tuning, the framework integrates an uncertainty-aware soft label aggregation mechanism that fuses conflicting annotations from human experts, vision language models, and pre-trained model predictions using the Dawid-Skene algorithm. Experimental validation on real-world ground test data demonstrates that RubNet achieves state-of-the-art performance with an F1-score of 93.26% and superior generalization capabilities across unseen sensor domains, while maintaining low computational complexity suitable for real-time deployment.

  • Open Access

    A Dual-Channel CBFA Network for Sleep Respiratory Event Detection Using SpO2 and Nasal Airflow

    by Yihao Wang, Jie Gao, Wenxiu Shu, Wei Wang

    Sleep Apnea-Hypopnea Syndrome (SAHS) is a common and severe sleep-related respiratory disorder that significantly impacts human health. Its main characteristics include the recurrent occurrence of apnea or hypopnea during sleep, leading to oxygen desaturation, disruption of sleep architecture, and various complications. Traditional diagnostic methods primarily rely on polysomnography (PSG), which, although highly valuable in clinical reference, is associated with issues such as complex testing procedures, poor real-time performance, and high costs of manual interpretation. To address these limitations, this paper proposes a deep learning-based SAHS detection method. The method utilizes blood oxygen saturation (SpO2) and nasal airflow signals from the UCD dataset as dual-channel inputs and constructs a composite model, CBFA, which integrates a one-dimensional convolutional neural network (1D-CNN), bidirectional long short-term memory (BiLSTM) network, and Forgetting Attention mechanism, for the automatic recognition and classification of sleep respiratory events. Experimental results demonstrate that the proposed model outperforms baseline methods in both classification accuracy and real-time detection capabilities, effectively extracting local abnormal features and temporal dynamic information. This method shows promising application potential and can provide technical support for the intelligent screening and auxiliary diagnosis of SAHS, with potential clinical value.

  • A Novel Simultaneous Calibration Method With Radius Constraint for 3-D Robot Measurement

    by Guangchao Tan, Wei Wang, Rui Guo, Xu Yang, Jinsong Du, Zhaojie Ju

    The utilization of 3D robot measurement systems, which integrate 2D profile scanners with industrial robots, is prevalent in manufacturing industries. Nonetheless, the measurement precision of such devices is often compromised due to inaccuracies in the robot kinematic parameters and the handeye matrix. This paper presents a novel simultaneous calibration methodology that concurrently resolves the issues related to robot kinematic parameters and the hand-eye relationship through the application of radius constraint. By integrating the hand-eye matrix into the robotic kinematic chain, an integrated model of the robot measurement system is formulated. Combining the spherical equation with the integrated kinematic model, a radius error model is established. The Jacobian matrix associated with the radius error is approximated using Taylor expansion. An iterative approach is proposed to identify the values of system parameters, which include both joint twists and hand-eye twist. Simulation results validate the effectiveness of the proposed algorithm, demonstrating its robustness to measurement noise. In practical applications, the maximum radius error was reduced from 4.480 mm to 1.721 mm, representing an accuracy improvement of 61.58%. Additionally, the average radius error decreased from 2.680 mm to 0.092 mm, indicating a substantial accuracy enhancement of 95.56%.

  • MMGT: Motion Mask Guided Two-Stage Network for Co-Speech Gesture Video Generation

    by Siyuan Wang, Jiawei Liu, Wei Wang, Yeying Jin, Jinsong Du, Zhi Han

    Co-Speech Gesture Video Generation aims to generate vivid speech videos from audio-driven still images, which is challenging due to the diversity of body parts in terms of motion amplitude, audio relevance, and detailed features. Relying solely on audio as the control signal often fails to capture large gesture movements in videos, resulting in more noticeable artifacts and distortions. Existing approaches typically address this issue by adding extra prior inputs, but this can limit the practical application of the task. Specifically, we propose a Motion Mask-Guided Two-Stage Network (MMGT) that uses audio, along with motion masks and pose videos generated from the audio signal, to jointly generate synchronized speech gesture videos. In the first stage, the Spatial Mask-Guided Audio2Pose Generation (SMGA) Network generates high-quality pose videos and motion masks from audio, effectively capturing large movements in key regions such as the face and gestures. In the second stage, we integrate Motion Masked Hierarchical Audio Attention (MM-HAA) into the Stabilized Diffusion Video Generation model, addressing limitations in fine-grained motion generation and region-specific detail control found in traditional methods. This ensures high-quality, detailed upper-body videos with accurate textures and motion. Evaluations demonstrate improvements in video quality, lip-sync, and hand gestures. The model and code are available at (https://github.com/SIA-IDE/MMGT).


  • Robust Hand—Eye Calibration With a Single Plane for 3-D Robot Measurement

    by Guangchao Tan, Jinsong Du, Rui Guo, Wei Wang, Zhaojie Ju

    Robots equipped with 2D laser profile scanners are widely utilized in industrial manufacturing processes. The hand-eye calibration is a critical aspect of robot 3D measurement and significantly influences the accuracy of the measured point cloud. This paper introduces a robust hand-eye calibration method that employing a single planar artifact in three steps. Firstly, the normal vector of the calibration plane is determined by utilizing points gathered through translation motions. Secondly, the rotation part is determined by the fact that the regression line of a single scan taken on a planar surface is orthogonal to the normal vector of the calibration plane. The translation vector is ultimately determined by leveraging the property that the line connecting the centers of two scans is perpendicular to the plane’s normal vector. The objective function is formulated using the Principal Component Analysis (PCA) technique and the Kronecker product. We propose a dual-constraint rotation objective function to enhance the robustness of the method against measurement noise. A closed-form solution is derived by employing the least squares method. The proposed methodology can directly calculate the accurate hand-eye matrix and exhibits a favorable tolerance for measurement noise originating from the scanner and deviations in robot pose. The effectiveness of the proposed algorithm has been confirmed through validation using both simulated data and real-world experiments.


  • A Lightweight Triple-Stream Network With Multisensor Fusion for Enhanced Few-Shot Learning Fault Diagnosis

    by Haotian Peng, Wei Wang, Jie Gao, Yu Wang, Jinsong Du

    The application of multiple sensors significantly enhances the accuracy of industrial fault diagnosis, but existing algorithms are structurally complex and rely heavily on extensive training data. To optimize the efficiency of diagnosis, this article proposes a lightweight time-frequency-statistical domain fusion network. The model comprises three data streams that analyze the time-domain, frequency-domain, and statistical features of vibration signals, employing an improved channel attention mechanism for weighted fusion. In addition, two model-agnostic few-shot enhancement strategies are introduced, aiming to improve accuracy where training samples are scarce by reducing signal sample variations and optimizing the distribution of signals in the feature space. By combining these techniques, the proposed method exhibits superior performance in few-shot learning on two datasets compared to other multisensor fusion methods, while also achieving higher computational speed. The results of this research are of significant importance in enhancing the fault diagnostic capabilities of multisensor systems in practical industrial applications.


  • Adversarial training of multi-scale channel attention network for enhanced robustness in bearing fault diagnosis

    by Haotian Peng, Jinsong Du, Jie Gao, Yu Wang, Wei Wang

    For bearing fault diagnosis problems in extremely noisy environments, this paper proposes an innovative universal adversarial training method. This method dynamically introduces noise into the training data, adaptively optimizing the model’s robustness. It applies to any neural network without incurring additional computational overhead in the reasoning process. Additionally, we introduce the multi-scale channel attention network (MSCAN). This network employs stacked convolutional kernels of varying sizes to extract features at different scales from the input signal. The incorporation of the channel attention mechanism allocates distinct weights to features of different scales, further enhancing the network’s representational capacity. Moreover, an automated machine learning-based automated tuning approach is employed to optimize the model training process, aiding in improving inference accuracy. Compared to existing designs, MSCAN exhibits exceptional accuracy. Through adversarial training, it maintains a 99.44% accuracy rate on the Case Western Reserve University dataset under strong −3 dB noise conditions. On the Paderborn University dataset at 0 dB, this adversarial training significantly improves the testing accuracy of various models by an average of 36.42%.