The Study Of The Face Recognition System Base On Embedded

Face recognition also say that automatic face recognition, Its relation technologywas used widely and payed more attention in the world in recent years. Facerecognition become an import branch and keep pace with fingerprint identification, irisrecognition and voice recognition.The products of the face recognition towards the miniaturization as well as thecomputer towards intelligence and miniaturization development. Because of thehardware limited of the embedded system that have not so much software such as thepersonal computer and it become the bottleneck of the face recognition products stepinto the mark.This paper first discusses the related algorithm for face recognition, and thealgorithm was improved to make its application in embedded system. In the system wecombined the skin color segmentation and Adaboost algorithm to build the Cascadeclassifier and achieve fast face detection. Then, we used the principal componentanalysis and pulse coupled neural network to extraction the face feature and put thisfeature as the input of the neural network classifier,at last we can get the result of theclassify. In the last of this paper, we use the S3C2440 processor and linux operatorsystem to build a face recognition system. I transplantation the opencv image processlibrary into the embed system and finished the face recognition completely.