FPGA-Based EMD Assist Block for Motion Detection in Critical Environments

Abstract

This study aims to carry out the implementation in FPGA (Field Programmable Gate Array) of a processing system capable of detecting and tracking moving vehicles without losing the system's reliability and efficiency of real-time application. Hence, for simple technique of tracking reach good results is proposed for removing shadows, an auxiliary processing based on insect vision system EMD (Elementary Motion Detector). The principle of this processing is based on cross-correlation and has been implemented in some work to get the direction of moving objects, however, this work is modified and implemented to auxiliary in removing shadows of objects detected using color, spatial and temporal information. The proposed auxiliary processing proves to be efficient in a video stream where the two most critical vehicles are selected with respect the cast shadow. The results of the implementation proposal are presented in terms of resource consumption and maximum operating frequency in a Spartan-6 FPGA-based development board. The experimental results are obtained in co-simulation system and show the effectiveness of the proposed system.