Infrared image recognition method and system based on visual processing and multi-feature fusion

Abstract

The invention provides an infrared image recognition method and system based on visual processing and multi-feature fusion, and the method comprises the steps: processing a collected infrared image through employing an image preprocessing technology, so as to filter the interference of a background and other factors on equipment recognition, carrying out the target detection of the infrared image through employing a target detection model based on VGG-Net multi-feature fusion, and carrying out the recognition of a target image. And the position of the power equipment in the infrared image is positioned, and the type of the positioned power equipment is identified by using the trained equipment classification model, so that the accuracy of infrared image equipment positioning and identification can be ensured, and the requirements of substation scene robot inspection can be met.

Description

technical field [0001] The invention belongs to the technical field of inspection robots and computer vision, and relates to an infrared image recognition method and system based on visual processing and multi-feature fusion. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] The research on the detection and fault diagnosis of electrical equipment using infrared thermal imaging technology has been vigorously carried out at home and abroad, and this technology has also been widely used in the existing substation inspection robot system. Infrared thermal imaging technology is a non-contact, passive measurement technology. It can detect and diagnose a large number of internal and external defects of power equipment, especially many faults that are inconvenient or impossible to detect by conventional test methods. However, according to the inv...

AI Technical Summary

Problems solved by technology

[0004] 1) Infrared thermal imaging technology reflects the temperature field information, which will cause serious deformation of the object shape

This technology cannot effectively locate and identify the infrared images of substation equipment, resulting in the inability to accurately map the corresponding relationship between temperature field information and electrical equipment and its various parts in the image;

[0005] 2) In order to identify substation equipment through infrared image intelligent recognition technology, we have accumulated a large amount of infrared image data resources, but traditional machine learning algorithms cannot effectively use these data resources, and cannot effectively troubleshoot substation equipment diagnosis;

The overall distribution of infrared image grayscale is low and concentrated, and due to the random interference brought by the surrounding environment to the infrared imaging process and the imperfection of the thermal imaging system itself, the signal-to-noise ratio and contrast of infrared images are relatively low

Therefore, the recognition of power equipment through infrared images has high requirements for target detection and positioning algorithms. Traditional visual recognition algorithms cannot meet the needs of equipment recognition in such complex scenarios.

Images