A design of object detection system in fog
Runze Huang, Jintao Wang · 2022 IEEE 6th Information Technology and Mechatronics Engineering Conference (ITOEC) · 2022
Images taken by intelligent monitoring equipment, unmanned driving, and intelligent transportation in foggy days have greatly affected the later object detection and information extraction. The main purpose of this research paper is to study and report how to implement an object detection system in foggy weather, and to implement a higher-precision Retinex algorithm on FPGA. The system uses Xilinx's ZYNQ 7 series development platform, the detection method is based on Tiny-YOLOv2. The Gaussian filter in Retinex algorithm uses a full variational design to enhance image quality and increase processing speed. The results prove that the accuracy of the YOLO object detection system is greatly improved after the retinex algorithm is processed, and it meets the requirements of fog object detection.