Identification on the burned forest area of Qipan Mountain using GF1-WFVsatellite image

International Conference on Manufacturing Technology, Materials and Chemical Engineering

材料科学;化学工业

Yu, W.Y.^1^2 ; Ji, R.P.^1^2 ; Feng, R.^1^2 ; Wu, J.W.^1^2 ; Zhang, Y.S.^1^2

Institute of Atmospheric Environment, China Meteorological Administration, Shenyang

110166, China^1

Key Laboratory of Agrometeorological Disasters, Liaoning Province, Shenyang

1101 66, China^2

关键词: After fires;    Burned areas;    Fire occurrences;    Forest fires;    High resolution satellites;    Overall accuracies;    Spatial resolution;    Temporal images;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/592/1/012164/pdf
DOI  :  10.1088/1757-899X/592/1/012164

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

In this study, GF1-WFV image was used with a spatial resolution of 16 m, which is higher than that of the commonly used MODIS, TM and other medium-high resolution satellites. Combined with the information extraction of pre-disaster forest and the ratio of NDVI in the two-temporal image before and after fire occurrence, the burned area, and its range after the "4.17" fire in Qipan Mountain in 2019 was analyzed. The results showed among the vegetable indices NDVI and EVI were better than GEMI and BAI in the extraction of burned forest area, while the radio of NDVI in two two-temporal was better than that of EVI. By using NDVI to extract forestland information before the fire occurrence and combining with NDVI ratio, it was most suitable to identify the burned area after a large range of forest fire in this area, with an overall accuracy of 92.2%.