QYQing Ye

Qing Ye

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Qing Ye has a total of 22 co-authors. Here are the top co-authors by number of shared publications:

  • Yongmei Zhang

    School of Information Engineering, North China University of Technology, Beijing, China

    5 shared publications
  • Chang Qu

    North China Univ Technol, Sch Informat Sci & Technol, Beijing 100144, Peoples R China

    2 shared publications
  • Rui Li

    North China Univ Technol, Sch Informat Engn, Beijing 100144, Peoples R China

    2 shared publications
  • Yongmei Zhang

    School of Information Science and Technology, North China University of Technology, Beijing, China

    2 shared publications
  • Zexian Tan

    System Operation Center, Bank of Beijing, Beijing 100033, China

    2 shared publications
  • Bo Huang

    School of Information Science and Technology, North China University of Technology, Beijing100144, China

    1 shared publication

Latest Publications

  • Open Access

    Helmet Detection Algorithm Based on Lightweight Model and Attention Mechanism

    by Bo Huang, Guorui Lv, Zichuan Guo, Wenqing Li, Qing Ye

    The detection algorithm can greatly improve the safety of construction workers under the background of construction sites and ensure the personal safety of construction workers. In this paper, the detection algorithm of helmet wearing is studied. YOLOv5 model is adopted as the core algorithm for personnel helmet wearing detection, and the lightweight model GHOST and attention mechanism CBAM are added to improve the original algorithm for construction scenarios. The lightweight model GHOST is a lightweight model used to reduce the computational load of convolutional neural networks. Attention mechanism CBAM used to amplify the receptive field of convolutional neural networks. Due to many dangerous factors, emergencies may occur at any time in the construction scene. Efficient and reliable personnel helmet wearing can further improve the operational efficiency and detection accuracy by using the attention mechanism. The algorithm has good performance in accuracy and robustness, and has high detection accuracy and good adaptability.

  • Open Access

    Dual-Channel Autoencoder with Key Region Feature Enhancement for Video Anomalous Event Detection

    by Qing Ye, Zihan Song, Yuqi Zhao, Yongmei Zhang

    Video anomaly event detection is crucial for analyzing surveillance videos. Existing methods have limitations: frame-level detection fails to remove background interference, and object-level methods overlook object-environment interaction. To address these issues, this paper proposes a novel video anomaly event detection algorithm based on a dual-channel autoencoder with key region feature enhancement. The goal is to preserve valuable information in the global context while focusing on regions with a high anomaly occurrence. Firstly, a key region extraction network is proposed to perform foreground segmentation on video frames, eliminating background redundancy. Secondly, a dual-channel autoencoder is designed to enhance the features of key regions, enabling the model to extract more representative features. Finally, channel attention modules are inserted between each deconvolution layer of the decoder to enhance the model’s perception and discrimination of valuable information. Compared to existing methods, our approach accurately locates and focuses on regions with a high anomaly occurrence, improving the accuracy of anomaly event detection. Extensive experiments are conducted on the UCSD ped2, CUHK Avenue, and SHTech Campus datasets, and the results validate the effectiveness of the proposed method.


  • Open Access

    A Ship Detection Method in Infrared Remote Sensing Images Based on Image Generation and Causal Inference

    by Yongmei Zhang, Ruiqi Li, Zhirong Du, Qing Ye

    Aiming at the problems of scarce public infrared ship data and the difficulty of obtaining them, a ship image generation method based on improved StyleGAN2 is proposed. The mapping network in StyleGAN2 is replaced with a Variational Auto-Encoder, enabling the generated latent variables to retain original image information while reducing computational complexity. This benefits the construction of the image. Additionally, a self-attention mechanism is introduced to capture dependency information between distant features, generating more detailed object representation. By reducing the number of input noises in the generator, the quality of the generated images is effectively enhanced. Experimental results show that the images generated by the proposed method closely resemble the structure, content and data distribution of the original real images, achieving a higher level of detail. Regarding ship detection methods based on deep learning, they often suffer from complex detection networks, numerous parameters, poor interpretability, and limited real-time performance. To address these issues, a lightweight multi-class ship detection method for infrared remote sensing images is designed. This method aims to improve real-time performance while maintaining accurate ship detection. Based on ship detection, an interpretable ship detection approach based on causal reasoning is presented. By integrating singular value decomposition with the Transformer architecture, the model focuses on causal ship features associated with labels in the images. This enhances the model’s robustness against non-causal information, such as background details, and improves its interpretability.


  • Open Access

    Human action recognition method based on Motion Excitation and Temporal Aggregation module

    by Qing Ye, Zexian Tan, Yongmei Zhang


  • Human interaction recognition method based on parallel multi-feature fusion network

    by Qing Ye, Haoxin Zhong, Chang Qu, Yongmei Zhang

    Human activity recognition is a key technology in intelligent video surveillance and an important research direction in the field of computer vision. However, the complexity of human interaction features and the differences in motion characteristics at different time periods have always existed. In this paper, a human interaction recognition algorithm based on parallel multi-feature fusion network is proposed. First of all, in view of the different amount of information provided by the different time periods of action, an improved time-phased video down sampling method based on Gaussian model is proposed. Second, the Inception module uses different scale convolution kernels for feature extraction. It can improve network performance and reduce the amount of network parameters at the same time. The ResNet module mitigates degradation problem due to increased depth of neural networks and achieves higher classification accuracy. The amount of information provided in the motion video in different stages of motion time is also different. Therefore, we combine the advantages of the Inception network and ResNet to extract feature information, and then we integrate the extracted features. After the extracted features are merged, the training is continued to realize parallel connection of the multi-feature neural network. In this paper, experiments are carried out on the UT dataset. Compared with the traditional activity recognition algorithm, this method can accomplish the recognition tasks of six kinds of interactive actions in a better way, and its accuracy rate reaches 88.9%.


  • Open Access

    Human motion recognition using three-dimensional skeleton model based on RGBD vision system

    by Qing Ye, Zexian Tan, Chang Qu, Li Zhang

    Human action recognition is one of the most important topics in computer vision. It has a lot of applications in intelligent monitoring, human-computer interaction, virtual reality, motion analysis and video annotation, etc. However, the effects of recognition are usually affected by human occlusion, self-occlusion, ambiguity and so on. In view of the above problems, this article describes an algorithm to recognize human motion using three-dimensional skeleton model based on RGBD vision system. Firstly, the human body three-dimensional skeleton model is established with the Kinect. Next, the normalized relative positions of skeleton joints and joint angles are adopted as features, and the action database containing the color image, the depth image, the skeletal image and the coordinates of the joints is built. Then, the features are trained and classified by multi-class support vector machine classifier. Finally, human motion recognitions are carried out. Experimental results indicate that the algorithm can recognize 12 kinds of actions behaved by different humans effectively.


  • Open Access

    Human Interaction Behavior Recognition based on Local Spatial - Temporal and Global Feature

    by Qing Ye, Rui Li, Xinran Guo

    At present, the human interaction behavior algorithm has received great attention, but how to effectively extract the interactive characteristics of human interaction and establish the complex model between multiple targets is still a difficult problem. Aiming at this problem, we proposed a human interaction recognition method based on local spatial - temporal and global feature. This method treats people who are interacting as a whole. Firstly, foreground human object is extracted by the inter-frame difference. Secondly, the optical flow and space-time points of interest are extracted. Considering the large and high dimension of the feature, the characteristics of each frame image are described by the gradient direction histogram HOG and the optical flow histogram HOF. Then, BP neural network is used to classify each feature of each frame. Considering the limitations of different features for interactive behavior recognition, BP neural network combined with classification fusion is used to obtain the final interaction behavior category. The algorithm tested in the UT interaction dataset, and the experimental results show that this method can recognize the interaction behaviors of hitting, hugging, kicking, pushing and handshaking. The proposed method can be applied to intelligent video surveillance.


  • Open Access

    Human Interaction Recognition Based on Whole-Individual Detection

    by Qing Ye, Haoxin Zhong, Chang Qu, Yongmei Zhang

    Human interaction recognition technology is a hot topic in the field of computer vision, and its application prospects are very extensive. At present, there are many difficulties in human interaction recognition such as the spatial complexity of human interaction, the differences in action characteristics at different time periods, and the complexity of interactive action features. The existence of these problems restricts the improvement of recognition accuracy. To investigate the differences in the action characteristics at different time periods, we propose an improved fusion time-phase feature of the Gaussian model to obtain video keyframes and remove the influence of a large amount of redundant information. Regarding the complexity of interactive action features, we propose a multi-feature fusion network algorithm based on parallel Inception and ResNet. This multi-feature fusion network not only reduces the network parameter quantity, but also improves the network performance; it alleviates the network degradation caused by the increase in network depth and obtains higher classification accuracy. For the spatial complexity of human interaction, we combined the whole video features with the individual video features, making full use of the feature information of the interactive video. A human interaction recognition algorithm based on whole–individual detection is proposed, where the whole video contains the global features of both sides of action, and the individual video contains the individual detail features of a single person. Making full use of the feature information of the whole video and individual videos is the main contribution of this paper to the field of human interaction recognition and the experimental results in the UT dataset (UT–interaction dataset) showed that the accuracy of this method was 91.7%.