Power operation site risk supervision intrusion video identification method

Abstract

The invention relates to a power operation site risk supervision intrusion video identification method, and belongs to an AI-based power grid production management safety operation management and control key technology, mainly aiming at identifying intrusion behaviors of a power operation site, being high in identification accuracy, and capable of realizing automatic monitoring and unattended operation. The power operation site risk supervision intrusion video identification method mainly comprises the following steps: step 1, initializing a background model; step 2, carrying out forward detection; step 3, updating the model; step 4, performing morphological filtering; step 5, acquiring a target area; step 6, performing regional module matching; and step 7, marking a non-staff invasion area. The power operation site risk supervision intrusion video identification method provided by the invention is a power operation field risk supervision means based on deep artificial intelligence, realizes high-efficiency supervision of the operation field, improves the supervision level, and reduces the operation safety risk.

Description

technical field [0001] The invention relates to the key technology of AI-based power grid production management safety operation management and control, in particular to a method for identifying intrusion video of risk supervision on the electric power operation site. Background technique [0002] There are various types of power grid operation sites. Different professionals are responsible for emergency repairs, tests, and maintenance at different job sites. Safety is no small matter, and safety is greater than the sky. Production safety risk management and control has become one of the most important tasks in the daily work of the power grid. However, in daily work, the work area is large, the work is complicated, and there are many people. However, the supervision and management personnel are limited. It may inadvertently lead to accidents, threatening the safety of personnel and the safety of power grid operation. [0003] At present, manual on-site inspection and remot...

AI Technical Summary

Problems solved by technology

The inspectors are limited, it is difficult to cover them all, they are often exhausted, and the audit efficiency is low

[0005] 2. Even if the inspectors arrive at the site, they come and go in a hurry, unable to carry out the safety inspection of the whole operation process

[0006] 3. Inspectors may not be able to strictly enforce the law

The large monitoring range of substation monitoring, the deployment of many monitoring points, and the random occurrence of abnormalities have brought a lot of work intensity to the substation staff

Images