Small target recognition method and device for power transmission system

Abstract

The invention discloses a small target recognition method and device for a power transmission system, and the method comprises the steps: carrying out the image enhancement of an image set with a target object, so as to improve the image resolution, carrying out the augmentation of the image set, increasing the number and diversity of images of the image set, adding a class label at a real position corresponding to the target object, using a feature extraction model constructed by a Darknet53 neural network as a feature extractor of a deep convolutional neural network, and establishing an image target recognition model through sample set training, so that the category of a target object in an image and corresponding position information can be quickly recognized.

Description

technical field [0001] The present application relates to the technical field of image recognition, in particular to a small target recognition method and device for a power transmission system. Background technique [0002] Today's society is increasingly dependent on electricity supply. At present, insulator identification methods are mainly traditional manual on-site inspection and UAV aerial photography inspection. Manual inspection technology has high recognition accuracy, but information collection is difficult and inefficient. UAVs are used for inspection of transmission lines, although information collection is convenient. , but it is still necessary to manually detect the collected images. Although this technology is more efficient than the traditional manual field detection technology, it will also be affected by related factors, such as the environmental impact of drones when shooting insulators and artificial long-term The detection of aerial images is prone to ...

AI Technical Summary

Problems solved by technology

At present, insulator identification methods are mainly traditional manual on-site inspection and UAV aerial photography inspection. Manual inspection technology has high recognition accuracy, but information collection is difficult and inefficient. UAVs are used for inspection of transmission lines, although information collection is convenient. , but it is still necessary to manually detect the collected images. Although this technology is more efficient than the traditional manual field detection technology, it will also be affected by related factors, such as the environmental impact of drones when shooting insulators and artificial long-term The detection of aerial images is prone to missed detection, etc.

[0003]At the same time, insulators and other electrical equipment are small in size and large in number. The insulators in the machine patrol image are small target detections. Under adverse conditions (such as rain, snow, frost and fog), it is difficult to quickly identify and locate small targets in the image

Images