项目作者: yash8005
项目描述 :
Human face identification has been a testing issue in the regions of picture preparing and patter acknowledgment. Another human face location calculation by crude Haar course calculation joined with the refreshed changes are to be examined.First, pictures of individuals are handled by a crude Haar course classifier, almost without wrong human face dismissal (low rate of false negative) yet with some wrong acknowledgment (false positive). Also, to dispose of these wrongly acknowledged non-human faces, a powerless classifier in light of face skin tint histogram coordinating is connected and a dominant part of non-human countenances are evacuated. Next, another frail classifier in view of eyes identification is attached and some leftover non-human appearances are resolved and dismissed. At last, a mouth recognition task is used to the rest of the non-human appearances and the false positive rate is additionally diminished.
高级语言: Python
项目地址: git://github.com/yash8005/Facial-Recognition-using-haar-cascade.git