Cheating behavior detection and early warning method and system

Abstract

The invention discloses a cheating behavior detection and early warning method and system, and the method comprises the steps: obtaining a monitoring start instruction, and extracting the picture information in monitoring video information according to the monitoring start instruction; combining the picture information with a preset convolutional neural network to obtain a plurality of priori boxes as pre-selected boxes; and matching the pre-selected boxes in combination with a preset cheating behavior atlas and a loss function, and sending early warning information when matching succeeds. Byautomatically monitoring the abnormal behaviors of the examinees in the examination room in real time, once it is monitored that the abnormal behaviors of the examinees belong to cheating behaviors, early warning information is sent out immediately in real time, accurate recognition of the cheating behaviors of the examinees is achieved, a large amount of manpower is saved, and meanwhile the workefficiency of monitoring the cheating behaviors of the examinees is greatly improved. The cheating behavior detection and early warning method and system can be widely applied to the technical field of image processing.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and system for detecting and warning cheating behaviors. Background technique [0002] Explanation of terms: [0003] SSD: (Single Shot MultiBox Detector) Single Shot MultiBox Detector. [0004] Faster RCNN: Faster Convolutional Neural Networks Based on Region Proposals. [0005] NMS: (Non Maximum Suppression) non-maximum suppression. [0006] Conv.: (Convolutional layer) convolutional layer. [0007] As an important link in teaching activities, examination is an important means to measure and identify the academic level of candidates. In the examination room, some test takers usually have a fluke mentality and cheat during the examination process in endless ways; cheating in the examination room not only seriously damages the rules of fair competition from the root, but also greatly weakens the value of the examination measurement function. At present, the...

AI Technical Summary

Problems solved by technology

However, the above two types of methods all rely on manual identification and judgment to varying degrees, and it is inevitable that there will be misjudgments or omissions. Observing the screen for a long time will not only easily cause visual fatigue to the staff, but also a large amount of redundant information in the video information will occupy a considerable part. hard disk resources, resulting in a waste of resources. In addition, it is difficult to effectively eliminate or reduce the occurrence of cheating behaviors through manual identification, which consumes a lot of manpower.

Images