Real-Time Multi-Target Detection System Based on FPGA Platform
Year: 2026
Type: conference-paper
Abstract: Aiming at the problem of missed detection and redetection of multi-target detection in video, a new detection method is proposed. Firstly, the adaptive median filtering algorithm is used to preprocess the image, and then the eight-direction Sobel edge detection operator is used to extract the edge features of the target. The edge features are applied to the three-frame difference algorithm and morphological processing method to form a collaborative mechanism of feature guidance and morphological optimization. In terms of hardware implementation, a real-time target detection system is built based on FPGA platform. The DDR cache is scheduled by ADMA transmission mode to realize low delay transmission and high efficiency operation. The results show that when processing 1024×768@30 frames of video, the system can detect multiple independent moving targets at the same time and meet the real-time processing requirements.
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7204887503
