一种基于FPGA的实时人脸检测系统设计

The Design of a Real-time Face Detection System Based on FPGA

针对Rein Lien Hsu提出的加入亮度补偿后的椭圆肤色模型在亮度较高和较低的区域容易误判的问题,提出了根据亮度值来分段选择椭圆长短半轴参数值的改进方法,MATLAB实验仿真表明改进后的算法比原始椭圆模型算法具有更高的适用性和准确性,搭建的FPGA系统能够满足人脸的实时检测需求。

In order to solve the problem of miscalculation in the high and low brightness areas generated by the elliptical skin model which is proposed by Rein Lien Hsu, this paper proposed a improved method by choosing the value of long and short elliptical half shaft according to the brightness value. MATLAB Simulation experiments show that this improved algorithm has higher applicability and accuracy than the original ellipse model algorithm, also the designed system based on FPGA could meet the need of real-time face detection.