|
|
|
| Research on Bionic Robotic Fish Object Detection Algorithm based on Improved YOLOv5 |
| WANG Ying1，SUN Kexin1，LIU Zhengang2，GAO Kangsheng1，LIU Qi1#br#
#br#
|
1. School of Information and Control Engineering，Jilin Institute of Chemical Technology，Jilin City 132022，China
2. Stauber (Hangzhou) Precision Mechanical Electronics Co., Ltd, Hangzhou 310018, China
|
|
|
|
|
Abstract To meet the needs of object detection for biomimetic robotic fish, a lightweight detection algorithm based on YOLOv5 is proposed to reduce algorithm complexity and improve accuracy. First, improvements are made to the YOLOv5s model by using GhostConv and C3Ghost modules to reduce the number of parameters and computational load. Second, CA and CoordConv modules are introduced to enhance feature extraction and target position perception capabilities, and soft NMS is used to reduce missed and false detections caused by traditional Non-Maximum Suppression (NMS). Additionally, MPDIoU is used to simplify similarity comparison, improving detection accuracy and recall rate. Finally, experimental results on the object detection dataset show that the improved YOLOv5 network is smaller in size and higher in accuracy, demonstrating the effectiveness and superiority of the proposed algorithm.
|
|
Published: 25 May 2024
|
|
|
|
|
| No related articles found! |
|
|
|
|
