Codebook hardware implementation on FPGA for background subtraction

Abstract

Object detection and tracking are main tasks in video surveillance systems. Extracting the background is an intensive task with high computational cost. This work proposes a hardware computing engine to perform background subtraction on low-cost field programmable gate arrays (FPGAs), focused on resource-limited environments. Our approach is based on the codebook algorithm and offers very low accuracy degradation. We have analyzed resource consumption and performance trade-offs in Spartan-3 FPGAs by Xilinx. In addition, an accuracy evaluation with standard benchmark sequences has been performed, obtaining better results than previous hardware approaches. The implementation is able to segment objects in sequences with resolution $$768\times 576$$ at 50 fps using a robust and accurate approach, and an estimated power consumption of 5.13 W.