On-duty personnel behavior identification method and system

Abstract

The invention provides an on-duty personnel behavior identification method and system, which utilizes an artificial intelligence algorithm and a big data analysis technology to improve the on-duty personnel supervision efficiency, and solves the problems of large workload, long time consumption and small coverage of manual inspection. The identification method comprises the following steps: acquiring image data, and inputting the image data into an optimized YOLOV3 model to extract a plurality of scene images with human-shaped contours so as to complete personnel detection; analyzing the detection result of the on-duty personnel to obtain the on-duty condition of the personnel; inputting the scene image into an SSD target detection model, and identifying whether a cigarette or a mobile phone is carried in the scene image; judging whether the behavior of the on-duty person violates an early warning strategy, the early warning strategy comprising a plurality of early warning levels, and sending early warning information to the user according to the corresponding early warning levels.

Description

technical field [0001] The invention relates to the technical field of visual recognition, in particular to a method and system for recognizing behaviors of on-the-job personnel. Background technique [0002] In some industries, monitoring the behavior of on-the-job personnel is a common management method. Its purpose is to prevent inappropriate behaviors of on-the-job personnel due to personal self-control and behavior habits, so as to ensure post discipline. Traditional job monitoring uses surveillance cameras combined with human inspections. The inspection time and scope are limited by various resources such as manpower, and there are problems of insufficient supervision and low supervision efficiency. [0003] In recent years, with the development of artificial intelligence technology and big data analysis technology, the visual algorithm based on deep learning has made great progress in the field of target detection and target recognition, but the use of visual recognit...

AI Technical Summary

Problems solved by technology

Traditional job monitoring uses surveillance cameras combined with human inspections. The inspection time and scope are limited by various resources such as manpower, and there are problems of insufficient supervision and low supervision efficiency.

[0003] In recent years, with the development of artificial intelligence technology and big data analysis technology, the visual algorithm based on deep learning has made great progress in the field of target detection and target recognition, but the use of visual recognition technology to monitor job behavior also has a high false alarm rate. , the types of early warning are limited, and the shortcomings of being easily affected by the site environment

Images