Application Of Face Recognition Algorithm Based On A-2D2PCA In Intelligent Classroom

At present,with the Chinese university campus expansion,the number of classrooms in colleges and universities also increased sharply.In addition,the rapid development of Internet and Internet of Things in our country in recent years,has promoted the construction of informationization in colleges and universities,and the construction of intelligent campus has begun to take shape.However,the development of the authority management in the classroom lags behind,the management of university classroom use rights is still managed manually by the university property management center,that is,staff of property management center go to the scene to open the classroom door or the teachers borrow the corresponding classroom key from the property management center.The management is not only inefficient,but with the increase in the number of university classrooms,the workload of university property management centers also increased dramatically.Therefore,designing an efficient and convenient classroom management system to manage the use rights of college classrooms is of great importance to improving the efficiency and reducing the workload of school faculty members,and it is of great significance to accelerating the process of building an intelligent campus.The extensive research in artificial intelligence and computer vision technologies has greatly accelerated the development of face recognition technology.The application of face recognition technology to the authority management in the classroom can bring great convenience to faculty and staff,but the traditional f'ace recognition algorithm is computationally intensive and occupies a large memory.It is difficult to run on embedded systems.The thesis aims to solve the problem of tedious work of the management of classroom use rights in classrooms of colleges and universities,and designed an intelligent classroom control system.A face recognition algorithm suitable for embedded system was proposed,that is,A-2D2PCA face recognition algorithm,which was applied to the intelligent classroom control system.The specific research work in the thesis is as follows:(1)By studying the development status of intelligent classroom at domestic and abroad,the thesis analyzed the shortcomings of the construction of intelligent classroom in the development of the authority management in the classroom,and put forward an intelligent control scheme.The scheme is teacher gain access to the classroom by identity authentication,the way of identity authentication is face recognition,RFID and fingerprint.Teachers can gain access to the classroom as long as they pass the two kinds of verification,the double verification method ensures the safety of the classroom.At the same time,teachers can choose the appropriate verification method according to their own situation,which provides convenience for the teachers to open the door.The property management center can remotely open or close the classroom door,and view the use of the classroom.(2)The process and method of face recognition were studied,and a face recognition algorithm A-2D2PCA was proposed,which is Advanced 2DPCA and PC A fusion algorithm.The A-2D2PCA face recognition algorithm constructs the covariance matrix directly from the original image.The algorithm uses the optimal projection matrix in the horizontal and vertical directions when calculating the eigenmatrix,the algorithm effectively overcomes the problem that 2DPCA can not eliminate row or column dependencies when processing based on a single direction.Meanwhile,A-2D2PCA algorithm uses PCA to extract the principal components from the feature matrix,and further reduces the number of features in the image,making the algorithm occupy less memory during the operation,which is beneficial to run on the embedded system.The A-2D2PCA algorithm proposed in this thesis was tested by ORL and Yale face database.The test results show that the recognition rate of A-2D2PCA is higher than that of 2DPCA and PCA algorithm when the number of face samples is more than 80.At the same time,A-2D2PCA algorithm extracts the features of face images twice.Compared with 2DPCA and PCA algorithms,A-2D2PCA algorithm uses less memory space and is more suitable for embedded systems with limited memory resources.Therefore,the thesis used A-2D2PCA algorithm as face recognition algorithm of the system.(3)The hardware circuit and the software program of the intelligent classroom control system were designed.The hardware circuit design of the system includes the selection of the main control chip,the design of the minimum system circuit,the face recognition module circuit,t the human-computer interaction circuit,and other peripheral modules circuit.The thesis completed the choice of communication mode between host computer and slave computer,the slave computers use CAN bus to communicate,and the host computers use Ethernet to communicate.The system software program design includes the slave computer program,the host computer program and the gateway program.The design of the slave computer program has completed the transplantation of the embedded operating system ?pC/OS-? and the graphical user interface library,and designed face recognition authentication program,fingerprint authentication program,human-computer interface program and other program.(4)The thesis completed the set-up of intelligent classroom control system hardware and software testing platform,and completed intelligent classroom control system test of slave computers,host computers and communication function.The test results showed that the A-2D2PCA algorithm proposed in this thesis can run steadily on the intelligent classroom control system,and the functions of the system meet the designed requirements,achieving the expect goal.