用FPGA实现卷积神经网络的人脸检测系统

FPGA Implementation of Face Detection System Based on Convolutional Neural Network

人脸检测技术被广泛应用在人脸识别和人脸追踪中,并且逐渐在嵌入式系统中应用。近年来,卷积神经网络使人脸检测技术在检测效果上取得了巨大的进步。在FPGA平台上,采用软硬件协同的方法实现了一个从数据采集到显示完整的基于卷积神经网络的人脸检测系统。实验结果表明,硬件数据格式采用16位定点数的情况下,本文的系统对人脸有良好的检测效果。

Face detection is widely used in face recognition and face tracking,and is applied in embedded system.In recent years,face detection has achieved a great improvement in detection result with the help of convolutional neural network(CNN).By software-hardware co-design,a CNN based face detection system is proposed,which is implemented from data acquisition to display on FPGA platform.The test results show that under the circumstance that the data precision is 16-bit fixed point,the system has a good face detection result.