An Efficient FPGA based Reconfigurable Architecture for Object Detection using Adaptive Threshold

Moving object detection is one of the feature extraction technique which is mainly used in movement detection, human activity analysis, airport safety, security and other such applications. In this paper, an efficient FPGA based reconfigurable architecture of moving object detection using Adaptive threshold method is proposed. This architecture is designed and optimized at hardware architectural level to get optimal hardware utilizations suitable for FPGA implementation. For this purpose, normal ripple carry adder is replaced by Kogge-Stone adder and Booth multipliers by Vedic multipliers. The Xilinx Spartan-6 FPGA is used to implement this architecture with the help of HDL programming. System generator tool is used to interface input video frames to check the real time performance of the proposed architecture. The hardware utilization comparison shows that the proposed architecture requires less hardware than existing methods.