Method and device for detecting illegal personnel departure based on image recognition

Abstract

The invention discloses a personnel illegal off-duty detection method and device based on image recognition. The method comprises the steps of performing image extraction on a monitoring video according to a preset time interval; performing feature extraction on the extracted image by using the trained convolutional neural network to obtain an image feature result; performing target scanning on the extracted image features in a sliding window mode by using a region suggestion network to obtain region suggested image features; taking image features according to the suggested area frame to perform ROI processing, and calculating the number of on-duty personnel according to the calculated foreground score of each target frame in the image; and when the number of the on-duty personnel is smaller than a preset target on-duty personnel number, judging that the personnel leave the duty illegally. According to the invention, the regional neural network and the convolutional neural network arecombined to accurately and effectively determine whether a person is illegally off duty in monitoring, so that security holes caused by illegal off-duty of the person can be effectively avoided.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a method and device for detecting illegal departure of personnel based on image recognition. Background technique [0002] In some scenarios, such as video monitoring room and other places, accidents in the video monitoring area are often not discovered in time due to personnel leaving their posts without authorization, resulting in failure to deal with the problem in a timely and effective manner, resulting in greater losses or safety issues. Therefore, for these scenarios, it is necessary to Real-time monitoring of personnel on duty to avoid security loopholes due to illegal departure of personnel. [0003] In this regard, the traditional method is to monitor images manually, but the method of manual video monitoring is inefficient, labor-intensive and difficult to guarantee accuracy. In the field of intelligent monitoring and identification, the existing technology ...

AI Technical Summary

Problems solved by technology

[0002] In some scenarios, such as video monitoring room and other places, accidents in the video monitoring area are often not discovered in time due to personnel leaving their posts without authorization, resulting in failure to deal with the problem in a timely and effective manner, resulting in greater losses or safety issues. Therefore, for these scenarios, it is necessary to Real-time monitoring of personnel on duty to avoid security loopholes due to illegal departure of personnel

[0003] In this regard, the traditional method is to monitor images manually, but the method of manual video monitoring is inefficient, labor-intensive and difficult to guarantee accuracy

In the field of intelligent monitoring and recognition, the existing technology uses the area change of the image area to judge whether a person is on duty or not, which is greatly affected by the environment and the accuracy rate is not high.

Images