With the development of UAV technology, the improper use of UAVs has seriously endangered public safety. In order to meet the needs of UAV intrusion detection, this paper proposes a method for UAV image detection using deep learning technology. This method performs lightweight processing in the YOLOv4 object detection model to obtain a faster detection speed model, and uses the CA attention mechanism module to replace the original SE attention mechanism module to build a new model (called CA-YOLOv4-L) with stronger object detection capabilities. Use the improved model to test the UAV pictures and the results show that: The improved model CA-YOLOv4-L achieves an average accuracy of 94.63% on UAV images, a detection speed of 39.66FPS, and a 34.8% reduction in parameters compared to the original model. The proposed method can effectively identify the UAV in the picture, can identify multiple UAV targets, and estimate the position of the UAV in the image.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    UAV Detection Based on Improved YOLOv4 Object Detection Model


    Contributors:
    Niu, Run (author) / Qu, Yi (author) / Wang, Zhe (author)


    Publication date :

    2021-09-01


    Size :

    459054 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Pedestrian Detection Algorithm Based on Improved YOLOv4

    Zheng, Ziheng / Ni, Chenhui / Zeng, Guolei | IEEE | 2023


    Detection Algorithm Based on Improved YOLOv4 for Pedestrian

    Kong, Wen / Qiao, Yidan / Wei, Ziqi | IEEE | 2022


    UAV Target Detection Algorithm Based on Improved YOLOv4

    Wang, Wenyue / Li, Jian / Zhang, Qi | Springer Verlag | 2022


    Maritime ship detection algorithm based on improved YOLOv4

    Chen, Wangcheng / Li, Yingshu / Wang, Xuemei et al. | British Library Conference Proceedings | 2023


    Improved YOLOv4-tiny network for pedestrian detection

    Fan, Pengbo / Chen, Tingzheng / Zhou, Zongtan et al. | IEEE | 2022