Design And Implementation Of An Embedded Face Recognition Software

Face recognition is a kind of computer technology which means using analysis and comparison of facial vision features information to identify people ,and it is one of the research focuses of Pattern Recognition and Artificial Intelligence.It is widely applied to access control system,judicial application,electronic commerce, video surveillance and other fields.Traditional high-performance automatic recognition system is mostly based on the application of PC platform which requires higter portability and applicability when put into practice. Especially in recent years,with the rise of intelligent home and the rapid development of embedded system,embedded intelligent video surveillance system becomes more and more important. Combined with the advantage of embedded devices, face detection, automatic identification system, the embedded intelligent video surveillance system can be widely applied to various types of access control systems, home security and other occasions.This article makes a study of embedded face recognition system based on OpenCV.The paper discusses the structure of OpenCV,structural features of Windows CE and transportats OpenCV the open source of intel in Windwos CE platform,then simplifies the code function after transplantation ;It is based on the study of classical Adaboost face detection algorithm and HMM face recognition algorithm and utilizes the powerful vision administration ability of OpenCV to put embedded face detection and face recognition into practice by PDX-057T tablet-PC.Moreover, it Combines the features of embedded face recognition system, trains the face classifier and also sets the threshold of face recognition .The system is simpler and more convenient than traditional face recognition system performed by PC.It is portable as well as easy to use.In addition, it fully takes advantage of embedded system and provides a reference for the home intelligent security system .