Power transmission line small sample temperature image amplification method based on image similarity

Abstract

The invention provides a power transmission line small sample temperature image amplification method based on image similarity, and the method comprises the steps: collecting an infrared image of a power transmission line, carrying out the preprocessing of the image, carrying out the positioning and segmentation of a temperature value through combining an accumulation pixel method and a vertical projection method, building 11 types of temperature image data sets according to the temperature characteristics, calculating the image similarity by combining projection accumulation and a histogram, improving the CGAN model, adopting the CGAN model improved based on the image similarity to carry out adversarial training after an original image data set is input, amplifying 11 types of temperature images, carrying out analysis from subjective and objective angles, and verifying the effect. The method has the advantages that the image is preprocessed through the binarization method of the self-adaptive threshold value, the binarization accuracy and reliability are improved, the image similarity is calculated in combination with projection accumulation and a histogram, and the loss function of the CGAN model is improved.

Description

technical field [0001] The invention relates to the technical field of small-sample image amplification, in particular to a small-sample temperature image amplification method for transmission lines based on image similarity. Background technique [0002] The transmission line is the key to the safe operation of the power system. The environment in which it is located is complex and prone to various faults, and has never caused abnormal temperature. The traditional detection method is manual detection, which is time-consuming, laborious and inefficient. The defect detection method based on infrared images can detect temperature anomalies in time, which is an effective means to reduce transmission line faults. Usually, a temperature measurement map will be generated on the right side of the infrared image, and the maximum and minimum temperature values ​​will be marked at the same time, so as to facilitate timely detection of faults, troubleshooting and statistics. In the p...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a small-sample temperature image amplification method for power transmission lines based on image similarity. By collecting infrared images of power transmission lines, the binarization method of adaptive threshold is used to remove complex background information; secondly, combined with the accumulation of pixels method and vertical projection method to locate and segment the temperature value, and establish a temperature image data set containing 11 types of transmission lines; combine projection accumulation and histogram to calculate image similarity to improve CGAN, and perform image amplification to overcome the problem of small transmission lines. The problem of unbalanced and low-quality samples in the sample temperature image data set improves the quality of image generation and helps to improve the accuracy of later detection

Images