Tailing pond semantic segmentation method based on photogrammetric data

Abstract

The invention discloses a tailings pond semantic segmentation method based on photogrammetric data, which comprises the following steps: collecting historical tailings pond data, including multi-view photos and spatial position data of a measurement area; performing data reconstruction on the collected historical tailings pond data to generate three-dimensional point cloud data, digital orthoimage data and digital elevation data; performing random downsampling on the generated three-dimensional point cloud data to generate an oblique photogrammetry point cloud data set; generating a tailings pond semantic segmentation model; and performing real-time semantic segmentation on the acquired to-be-analyzed tailings pond photogrammetry data, and generating a photogrammetry image with a tailings pond semantic segmentation result in real time. According to the method, the point cloud data and the DOM data produced by oblique photogrammetry are combined, semantic segmentation is performed on the tailing pond based on the deep learning model, the tailing pond land type can be accurately and efficiently segmented, the method is simple, and the cost is low.

Description

technical field [0001] The invention belongs to the field of image data processing, in particular to a method for semantic segmentation of tailings ponds based on photogrammetry data. Background technique [0002] Tailings pond is a place to store tailings or other industrial wastes discharged from metal or non-metal mines after ore sorting. It is an essential infrastructure and environmental protection project for mining enterprises. The high potential energy of the tailings pond makes the mine operation site have the potential danger of man-made debris flow. Once an accident occurs, it is very easy to cause a dam failure and cause major safety accidents. [0003] Semantic segmentation of tailings pond physical information (such as initial dam, accumulation dam, water surface, dry beach, etc.) is an important basis for analyzing the status quo of tailings ponds. Important support for indicators. The traditional analysis method of status quo of tailings ponds usually reli...

AI Technical Summary

Problems solved by technology

The traditional analysis method of status quo of tailings ponds usually relies on manual labor to obtain the geometric information of tailings ponds through manual on-site inspection or manual use of measurement tools; however, the manual operation mode is inefficient and the data integrity is not strong

Images