Automatic labeling method for railway external environment risk source sample

Abstract

The invention discloses an automatic labeling method for a railway external environment risk source sample. The method comprises the steps that 1, DOM data and DLG collection result data in a regionalrange are sorted and subjected to quality inspection; 2, attention risk source target element information is extracted and screened; 3, vector polygon element topology errors are checked and corrected; 4, color uniformization processing is performed on the digital ortho-image; 5, a regular tile grid of the digital orthoimage is calculated and sample data is cut and output; 6, sample labeling range space analysis operation is performed; 7, the spatial position of the positive sample vector polygon is corrected; 8, a labeling result and an intermediate result are generated; 9, labeled data is checked and locally repaired; and 10, sample library data arrangement and unified output are performed. According to the method, the deep learning sample library is quickly generated from the large-format digital orthoimage data in batches, so that the sample labeling efficiency is effectively improved, the offset of vector acquisition and image representation ground object boundaries can be automatically corrected according to the sample data, and the sample labeling precision is improved.

Description

technical field [0001] The invention relates to the field of off-road environment improvement for railway operation management, in particular to a method for automatically marking external environmental risk source samples along a railway by using high-resolution remote sensing aerial survey data and vector collection result data. Background technique [0002] The operating mileage of my country's railways reached 140,000 kilometers by the end of July 2020, of which the mileage of high-speed railways exceeded 36,000 kilometers. A large number of human activities around the railways have created a large number of security threats to the safe operation of railways. Illegal buildings, mainly colored steel houses, have repeatedly caused railway traffic safety accidents. The complexity of the external environment of the railway has brought great pressure and challenges to the work of railway safety operation and rectification. The operation mode of manual on-site inspection has d...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the shortcomings of existing manual labeling methods for establishing a sample library, the present invention provides a method for automatically labeling samples of risk sources in the external environment of railways, which can be processed in batches on a large scale and has a high degree of automation and sample labeling efficiency

Images