Video intelligent analysis auxiliary inspection and abnormity warning method

Abstract

The invention provides a video intelligent analysis auxiliary inspection and anomaly warning method, which comprises the following steps of: classifying video image data, and performing anomaly labeling on images in a video image library based on an image description algorithm; constructing a training set according to the generated basic alarm record data, and constructing a learning model for alarm convergence; and setting an alarm threshold index, and performing classification output on the video image data of the video image library, the classification including effective faults, faults tobe detected and no faults. According to the invention, an alarm can be given in time when an abnormal accident occurs; screening is timely and convenient, and laws can depend on the screening; according to a searchable and system research equipment state polling and screening method, alarm threshold indexes are formulated, corresponding upper and lower limits are formulated according to alarm types and alarm levels, accidents can be selectively checked and alarm events can be reported according to filtering conditions, false alarms are reduced, missed alarms are reduced, and the reliability ofthe system can be greatly improved.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular, the invention relates to a video intelligent analysis auxiliary inspection and abnormal alarm method. Background technique [0002] At present, my country's transmission lines are still dominated by overhead lines. High-voltage and ultra-high-voltage transmission lines are distributed in many points and wide areas. The terrain is complex and the natural environment is harsh. In mountainous areas, due to soil erosion, rain, and debris flow Damage to the foundation of power transmission lines, especially power facilities such as towers, will lead to the toppling of towers or equipment damage, directly damaging the transmission lines. Power lines and tower accessories have been exposed to the wild for a long time, and are affected by continuous mechanical tension, lightning flashover, material aging, human-made and other factors, resulting in tower collapse, broken strands, wea...

AI Technical Summary

Problems solved by technology

For fault points with obvious discharge traces on the ground or transmission line equipment, it is easy to be found manually, but for cases where the discharge point is not obvious or there is a line-of-sight or viewing angle limit, it is almost impossible to find it manually through a high-power telescope on the ground. Finding the fault point by stepping on the tower or even routing the live line not only greatly increases the workload of the line operation and maintenance personnel, but also poses a safety threat to the live work personnel

Images