基于OPENCV具有眨眼侦测功能的人脸检测系统

Design of face detection system with the blink detection function based on OpenCV

随着社会智能化程度的提高,视频图像凭借其直观优势将逐渐取代文字的地位。本文在基于OPENCV的基础上,提出一个具有眨眼侦测功能的人脸检测系统,其中利用相对比较成熟的开源计算机视觉库opencv开源函式库对摄像头实时输入的图像进行采集,对采集图像做前期处理;Adaboost算法由于它特殊的算法模式,可以进行快速的目标检测,因此采用Adaboost算法做人眼定位检测,本文提出的眨眼侦测功能可以为系统提供防伪功能,为人脸识别技术提供一道安全的屏障,实验结果表明,本系统实验了人脸检测,人眼定位,眨眼侦测等功能,检测效率高,效果好。

The face detection system is more and more widely used in public because of its feature of high integration,high performance,high speed,low cast and stability.This paper presents how we compromise Face detection module with the blink detection function based on the OPENCV source.the camera is necessary which can be invoke by the OPENCV2.0.Each frame of camera wil be send to the face module to test the possibility existence of a human face.After the detection of human face,the system wil go further to detect the existence of the eye and locate the accurate position within the the algorithm of Adaboost.