Implementation of Motion Object Detection using BMNCKFCM on FPGA and ASIC

ABSTRACT Real time detection of moving objects was extremely crucial and fundamental step in the extraction of information concerning objects in motion and to stable the practical areas, such as tracking, classification, recognition, and so on. It is due to various factors like less competence of motion detection algorithm; increased hardware architecture, and poor memory access reduction schemes. To make efficient hardware architecture and increased hardware efficiency, of motion object segmentation, the proposed method used Background Modelling Neighborhood Coefficient Kernel based Fuzzy-C-Means (BMNCKFCM) algorithm was used in this paper. By the use of this algorithm keep a strong motion image to address variations on environmental changing provisions and utilized to eliminate the background interference information and separate the moving object from it. In fact, the consistency with which potential foreground objects in movement can be identified, directly impacts on the efficiency and performance level attained by subsequent processing steps of object recognition. The proposed algorithm has immense ability of anti-interference and maintains more accurate rate detection at the same time. The effectiveness of the presented algorithm for movement detection is displayed in a simulation environment and the assessment results of hardware like FPGA and ASIC are stated in this paper. In the view of Hardware realization of the motion object detection algorithms on FPGA and ASIC occupies more space, time and power consuming. To address all the issue of hardware architecture and observe effects from parameter settings in addition to fixed point quantization model can be performed with the help of FPGA and ASIC platform.