Research And Design Of Attendance Software System Based On Face Recognition

Face recognition is one of the most challenging research topics in the field of computer vision and artificial intelligence. With the development of society, people have higher requirements for the accuracy, rapidity and safety of the identification. Biometric technology has been developed rapidly because of its own advantages. Compared with other biometrics such as fingerprint recognition, voice recognition and iris recognition, face recognition technology has great development potential in the field of identity recognition, automatic monitoring, human-computer interaction and so on. Because it has the advantages of directness, Friendliness, non aggression and high stability. Face recognition technology mainly includes two parts:face detection and face recognition (matching). Face detection is the process of detecting the position of human face from static images or dynamic videos. Face recognition is the process that recognizes unknown faces by matching it with known faces in the database based on the similarity of extracted features.The significance of face recognition technology and the development of technology are firstly introduced in the paper.The main application scenarios and the current research difficulties are discussed. Secondly, the image gray, histogram equalization, spatial filtering, image sharpening and image geometry normalization are analyzed in the aspects of image preprocessing. AdaBoost face detection algorithm is used in face detection.This paper deeply researches the theory basis of AdaBoost algorithm, the Haar matrix characteristic and the integral graph.The real-time detection process are realized. The requirements of real-time and high accuracy of the attendance software system are met. Then, the main algorithms of face recognition are studied and compared. The face recognition algorithm based on PCA feature, Fisherfaces and LBP are studied and analyzed. The basic principle and the experimental procedure of the three algorithms are analyzed respectively. At the same time, the feasibility of the algorithm is verified through MATLAB. The face recognition algorithm based on LBP is finally selected as the face recognition algorithm of the attendance software system through the comparison of the three algorithms in four aspects of the training time, storage space, recognition time and recognition accuracy. Finally, a prototype system based on face recognition is designed and developed. Using software engineering, the whole process can be divided into feasibility study, requirement analysis, overall design, outline design and implementation. The performance of the prototype system is tested by using the ORL standard face database and self built face database. The experimental results show that the prototype system achieves the design goal of the system in the accuracy and real-time performance.