Method for electric transmission line image processing under complicated background based on image classification

Abstract

The invention discloses a method for electric transmission line image processing under a complicated background based on image classification and belongs to the technical field of picture processing.Electric transmission line image characteristic extraction under the complicated background is completed through image characteristic variable statistics, image classification, primary filtering, contrast enhancement, secondary filtering, edge defection and edge optimization. According to the method, images are classified according to variable values in different color spaces of the electric transmission line images, the pertinence is strong, and a processing effect is good. Linear transformation is conducted on average image gray values and threshold values calculated through a maximum inter-cluster variance method to determine histogram equalized segmented intervals, stretching intervals and parameters of a canny operator, the contrast enhancement and edge characteristic extraction of the images are better completed, and the applicability is good. An edge optimization method is provided, and the influence of non-target edges is removed to the most degree. The method can perform self-adaptive image processing according to image characteristics, is high in reliability and has the great significance in intelligent online monitoring of electric transmission lines.

Description

technical field [0001] The patent of the present invention relates to the technical field of image processing, in particular to an image classification-based image processing method for transmission lines under complex backgrounds. Background technique [0002] The transmission line is an important part of the power system. It is responsible for the task of transmitting and distributing electric energy. Abnormal conditions such as icing, galloping, disconnection, and excessive sag of the transmission wire can easily lead to transmission line failure and seriously affect the power system. Safe and stable operation, resulting in huge economic losses. Therefore, it is of great significance to the safe operation of the power grid to do a good job in the monitoring and prevention of transmission lines, to detect and deal with transmission line faults in time. [0003] At present, the monitoring of high-voltage transmission lines is mainly manual inspection, and information is ob...

AI Technical Summary

Problems solved by technology

At present, there are very few patents in this area in China. For example, the Chinese patent "Measurement Method for Icing Thickness of Transmission Line Insulators Based on Video Difference Analysis" (application number: 201110173506.9) and other types of patents have proposed the use of image processing methods to analyze transmission lines. It has the advantages of high efficiency and strong applicability, but most of the current domestic patents only discuss image processing methods for several images of the same type, without considering the influence of different image backgrounds on the image processing effect of transmission lines , it is difficult to correctly analyze transmission line images with different background types, which has very large limitations and cannot realize highly intelligent online monitoring of transmission lines

Images