Mine well wall detection method under low-illumination condition based on convolutional neural network

Abstract

The invention discloses a mine well wall detection method under a low-illumination condition based on a convolutional neural network. The mine well wall detection method comprises a training stage andan online detection stage. The training stage comprises the following steps: constructing an image decomposition network and an image detection network, and respectively training the image decomposition network and the image detection network by acquiring an image decomposition data set and an image detection data set, wherein the online detection stage is composed of image decomposition, image enhancement and image detection. An on-site mine well wall image is obtained and decomposed into a reflection image and an illumination image through a trained image decomposition network, then brightness enhancement is conducted on the well wall image through image enhancement, finally the well wall image is detected through a trained image detection network, and mine well wall state detection under the low illumination condition is achieved. According to the invention, the accuracy of mine well wall detection is improved, the operation cost of mine well wall detection is reduced, and the safety is improved.

Description

technical field [0001] The invention relates to the field of mine shaft wall detection methods, in particular to a mine shaft wall detection method under low illumination conditions based on a convolutional neural network. Background technique [0002] The topography and environment in most mine areas in my country are relatively harsh, and the strength of the rock mass of the mine is insufficient. In addition, the adverse effects caused by the consolidation of the stratum and the rise of the groundwater level, etc., the mine shaft wall often produces a large internal stress. When the stress is greater than the shaft wall structure When the ultimate strength is reached, there will be shaft wall damage and mine collapse accidents. In order to timely and accurately detect the damage of the mine shaft wall and reduce the safety hazards of mine production, it is necessary to accurately and efficiently detect the mine shaft wall. [0003] At present, the detection of mine shaft wa...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a mine shaft wall detection method under low illumination conditions based on a convolutional neural network to solve the problems of poor detection effect, low accuracy and long time-consuming in the manual detection of mine shaft wall conditions in the prior art

Images