Open AccessFormation Lithology Identification Technology along Railway in Complex and Dangerous Mountainous Area Based on Hyperspectral
Author(s) -
Jianhua Cai,
Limin Wang,
Y. L. Huang,
Dayang Chen,
Huan Liu
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/780/4/042049
Subject(s) - hyperspectral imaging , remote sensing , terrain , lithology , thematic map , feature (linguistics) , geology , computer science , environmental science , cartography , geography , paleontology , linguistics , philosophy
In view of the difficulties during implementation of conventional survey methods for railway projects in complex and dangerous mountainous areas, the section from Maoxian to Songpan in Chengdu-Lanzhou Railway with complex terrain and bad climate is selected as the research area to extract hyperspectral remote sensing technology content of the research area. Based on the research of indoor hyperspectral image processing, outdoor rock sample collection and indoor spectrum analysis, establish a suitable feature extraction and lithologic recognition classification method according to the unique spectrum integration feature of hyperspectral remote sensing image. Use the visible short wave infrared hyperspectral data of the latest domestic Gaofen-5 satellite, carry out the lithologic plotting experiment in the key areas of the section from Maoxian to Songpan in Chengdu-Lanzhou Railway, and make accuracy evaluation by selecting a certain way for the mapping results. It provides important thematic information for regional lithologic investigation, and provides strong technical support for selection, survey and design of Chengdu-Lanzhou Railway construction, which is helpful for similar projects.