Power equipment online sensing and focusing algorithm based on deep learning

Abstract

The invention provides a power equipment online sensing and focusing algorithm based on deep learning, and the algorithm comprises the following steps: controlling an unmanned plane to carry out the fixed-point cruise, carrying out the online detection and analysis of a collected image, and adjusting the shooting angle and focal length of a camera according to an analysis result; according to thepower equipment online perception and focusing algorithm based on deep learning, the position of typical power equipment in an image is captured by using a deep learning means; whether the camera focal length is set properly is analyzed by using a focusing evaluation function, the shooting angle and focal length of the camera are adjusted by integrating the two information to obtain a clear imageof the equipment in the image center, the whole process of the algorithm is completed and adjusted online by the unmanned aerial vehicle, manual intervention is not needed, and the efficiency and quality of inspection work can be improved.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and specifically relates to an online perception and focusing algorithm for electric power equipment based on deep learning. Background technique [0002] Large-scale power lines need to cross mountains and rivers. The traditional way of manpower line inspection is costly and inefficient, and it is difficult to meet the current needs. However, due to the influence of weather, lighting and experience of drone operators, the quality of drone patrol picture data is uneven, and the collected images often have inaccurate focus, unreasonable shooting angles, and blurred photos, etc. The problem is that the unstable quality of the inspection pictures has caused two main problems. One is that frequent re-shooting and re-shooting of the same location is required during the inspection process, which increases the workload of the inspectors and reduces work efficiency. The second is that i...

AI Technical Summary

Problems solved by technology

[0002] Large-scale power lines need to cross mountains and rivers. The traditional way of manpower line inspection is costly and inefficient, and it is difficult to meet the current needs. However, due to the influence of weather, lighting and experience of drone operators, the quality of drone patrol picture data is uneven, and the collected images often have inaccurate focus, unreasonable shooting angles, and blurred photos, etc. The problem is that the unstable quality of the inspection pictures has caused two main problems. One is that frequent re-shooting and re-shooting of the same location is required during the inspection process, which increases the workload of the inspectors and reduces work efficiency. The second is that in the process of massive image data processing in the later period, photos that do not meet the image quality requirements need to be eliminated, which increases the difficulty of manual or automatic data processing in the later period. Based on this, the present invention proposes an online perception of power equipment based on deep learning and focus algorithm

Images