Video monitoring dynamic irregular multi-supervision area discrimination method based on template matching

Abstract

The invention relates to image processing, computer vision and deep learning, and belongs to the field of intelligent security and protection. The method comprises the steps of template generation, target detection and region discrimination. The target detection module is used for processing an input video monitoring image, detecting coordinates of a target in the image and inputting coordinate information into the region discrimination module, and the region discrimination module is used for generating a generated supervision region template through the template according to the coordinate information of the target and discriminating a supervision level of a supervision region where the target is located; according to the method, the supervision level of the supervision area where the target is located can be judged in a video monitoring scene, illegal intrusion of the supervision area is actively prevented, and safety guarantee is provided. And the method is convenient to implement and easy to popularize and apply.

Description

technical field [0001] The invention relates to image processing, computer vision and deep learning, and belongs to the field of intelligent security. Background technique [0002] At present, in industrial production, there are often some high-risk areas where personnel are prohibited from approaching, such as belt corridors, high-voltage transformer boxes, high-temperature boilers, etc. In order to ensure the safety of personnel, enterprises usually use video surveillance for safety supervision. Real-time monitoring, but there are shortcomings such as easy distraction of video surveillance personnel and lagging hazard response. [0003] With the rapid development of deep learning technology, computer vision technology, and image processing technology, video surveillance intelligent processing methods have gradually emerged, which can judge whether there is object movement or simple target recognition in video surveillance, but in industrial environments, video surveillance...

AI Technical Summary

Problems solved by technology

[0002] At present, in industrial production, there are often some high-risk areas where personnel are prohibited from approaching, such as belt corridors, high-voltage transformer boxes, high-temperature boilers, etc. In order to ensure the safety of personnel, enterprises usually use video surveillance for safety supervision. Real-time monitoring, but there are disadvantages such as easy distraction of video surveillance watchers and lagging hazard response

[0003] With the rapid development of deep learning technology, computer vision technology, and image processing technology, video surveillance intelligent processing methods have gradually emerged, which can judge whether there is object movement or simple target recognition in video surveillance, but in industrial environments, video surveillance scenes are complicated. , the existing methods cannot meet the requirements

Images