Application Research Of Pose Detection With Perspective Transformation In Classroom

With the continuous development of the social economy,the development of education has become a hot topic of concern.The atmosphere of the classroom determines the quality of classroom teaching,and the behavior of students in the classroom reflects their listening status.At present,the number of students in schools continues to expand,and the evaluation of students’ behavior habits and learning attitudes has become a major difficulty.Monitoring equipment.However,in the context of the current educational environment,the combination of cameras and video recorders leads to a lack of intelligence and has a limited effect.In view of this current situation,this paper studies the classroom posture detection system,combines the real classroom environment,uses the Open Pose human posture recognition model as the key point detection framework,and proposes a posture detection method based on incremental learning to adapt to classroom human posture detection.Aiming at the distortion of classroom video images,a wide-angle image correction method based on perspective transformation is proposed.The main research contents are as follows:(1)Research on pose detection method based on incremental learning,aiming at adapting to the problem of pose detection in the classroom background,establish a gesture detection model based on incremental learning.Three classroom states are detected and judged.(2)Establish a wide-angle image correction model based on perspective transformation,use the perspective mapping method for the distortion generated by the wide-angle image,extract the coordinates of the corner points in the image,map the coordinate points according to the model,and use polynomial curve fitting to obtain the corrected image.,in order to improve the pose detection and evaluation standards of the students in the front and rear rows.(3)To build a system based on the research content,from the point of view of functions and needs,RK3288 is selected as the central processing unit,and the image acquisition unit is selected from Hikvision cameras,combined with NVR to realize multi-channel camera management.The MQTT transmission protocol is used to send and receive data information remotely,and the pose detection system under the classroom background is designed and implemented on the QT development platform,so that the system can adapt to various complex classroom backgrounds.Finally,the overall debugging and result analysis of the system are carried out.The test results show that the system achieves the expected effect in response speed and has certain practical value.