Recognition of students combining features of Zernike moment and optical flow

Due to practical application demand of automatic recorded education system, it is necessary to detect and recognize students' actions in the classrooms. This paper proposes a method combining with Zernike moment of motion history image and computing optical flow to recognize class actions of students, which mainly identifying three actions of hands up, stand up and sit down. Zernike moment, good sampling and strong ability to resist noise, can describe the shape of motion history image. Lucas-Kanade optical flow method can describe the direction and speed of the movement. Based on video database of Student-Behaviors captured in the paper and combining two ways to describe characteristics of the actions, experiment results show that the proposed method recognizes these three actions effectively, and has a certain robustness.