A power transmission line background target classification system based on an unsupervised SDAE network

Abstract

The invention discloses a power transmission line background target classification system based on an unsupervised SDAE network. The power transmission line background target classification system comprises an unmanned aerial vehicle video information acquisition module, a neural network pre-training module, a saliency target detection module and a background target classification and calibrationmodule. In the power transmission line background target classification process, an unmanned aerial vehicle video acquisition module acquires power transmission line background area image informationby using unmanned aerial vehicle-mounted video acquisition equipment; the neural network preprocessing module carries out DBN network training on the image; A saliency detection initial parameter is acquired, a saliency target detection module acquires a reconstructed image by using an SDAE network and is matched with the original image to acquire a saliency target calibration image, and a background target classification and calibration module inputs a multi-scale feature vector into a DBN network to train a rapid classification target object. The power transmission line background target classification system provided by the invention can accurately classify the power transmission line background targets, and the analysis result is stable and objective.

Description

technical field [0001] The invention relates to an unsupervised SDAE network-based background target classification system for power transmission lines, belonging to the fields of image segmentation and target detection and classification. Background technique [0002] With the continuous development of my country's electric power industry, the transmission line network, as an important part of the national grid, is an important national infrastructure, and its number and scale are constantly expanding. In rural areas, residents often illegally build houses with high floors along the transmission line, which seriously affects their own life and property safety and the normal operation of the transmission line. In urban areas, due to the complex regional planning and large population, the requirements for daily inspection are higher, and it is necessary to detect possible building safety hazards in time. In sparsely populated areas, there are mostly open spaces and trees alo...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is: accurately and quickly find salient objects in the complex background of power line images, and be able to classify common background objects

Images