TITLE:
Study on UAV Image Extraction of Surface Crack Information Technology in Goaf
AUTHORS:
Shihua Geng, Wenjun Zhang, Shuojuan Tian, Yingkun Zi, Junyi Miao, Rui Shen
KEYWORDS:
UAV Image, Object Oriented Information Extraction, Goaf, Surface Crack
JOURNAL NAME:
Open Access Library Journal,
Vol.9 No.4,
April
26,
2022
ABSTRACT: The problem of extracting surface crack information in mining area is studied. The unmanned aerial vehicle (UAV) image is used as the data source for multi-scale segmentation. The rule set is constructed based on the spectral and shape features of surface cracks. The object-oriented extraction of crack information is carried out by using the principle of fuzzy classification, and the surface crack information in mining area is successfully extracted. The extraction accuracy of surface cracks is evaluated by five indexes: automatic extraction number, extraction rate, automatic extraction total length, length extraction rate and position overlap rate. Through statistical analysis, the overall extraction rate of surface cracks by object-oriented technology reaches more than 80%, which proves that this method has high accuracy. This paper aims to explore an effective fracture extraction method suitable for the surface conditions of Jinchuan copper nickel mine. The feasibility of applying UAV technology to visual monitoring of surface cracks in goaf is confirmed.