Examination cheating identification algorithm based on image analysis

Abstract

The invention relates to an examination cheating identification algorithm based on image analysis. A device for implementing the algorithm comprises a bracket arranged on each desk, a camera and an LED lamp installed on the bracket. A processor is arranged in the bracket and is connected with the camera and the LED lamp; the device comprises a processor and a network module connected with the processor, a data link between the processor and a remote management server is built through the network module, and the processor sets an examination cheating identification algorithm. The algorithm comprises the following steps: S1, collecting a face image fi (x, y) of a testee; S2, extracting a face rectangular region from the face image fi (x, y); S3, extracting positions of a left eye and a righteye from a face rectangular region of the face image fi (x, y); S4, calculating an intermediate point M coordinate between the left eye and the right eye; S5, calculating position parameter coefficients Rx and Ry of the intermediate point M in the rectangular area of the human face; S6, when the absolute value of Rx is smaller than Kx and Ry is smaller than Ky, the human face faces right ahead and is in a normal answering state; otherwise, the human face faces left or right and is in a possible cheating state.

Description

technical field [0001] The invention relates to an examination cheating recognition algorithm based on image analysis, and belongs to the technical field of image-based pattern recognition. Background technique [0002] The job of a teacher is to preach the word, receive karma, and solve doubts. However, in addition to teaching and educating people, teachers currently have to do a lot of hard work, such as correcting a large number of homework and invigilating exams. In particular, the invigilation part makes the teachers feel particularly boring and hard: each examination room usually assigns at least two invigilators. During the exam, the invigilators must stand or walk for a long time, and keep observation around, and answer the questions of the candidates at any time. [0003] With the substantial improvement of camera imaging quality and the development of image processing algorithms, computer vision can achieve the ability of human behavior analysis and recognition. H...

AI Technical Summary

Problems solved by technology

[0002] The job of a teacher is to preach the word, receive karma, and solve doubts. However, in addition to teaching and educating people, teachers currently have to do a lot of hard work, such as correcting a large number of homework, invigilating exams, etc.

Images