Study on scale effect and decision fusion of coal mining collapse seeper extraction

• 1、China University of Mining and Technology, Key Laboratory for Land Environment and Disaster Monitoring of SBSM, Xuzhou 221116

Abstract：Using remote sensing imagery to extract the coal mining collapse seeper is an effective way of geological environment monitoring and investigation. Extracting coal mining collapse seeper from remotely sensed images are often faced with some problems in application, such as the choose of optimal scale, the improve of recognition and extraction precision et al. Based on the thought of scale effect and decision fusion, using the ALOS multi-spectral data, ASTER VNIR data and CBERS multi-spectral data to extract the mining collapse seeper in study area. According to the mining area, the experimental results show that: for the single data source, the optimal scale extraction is the ASTER VNIR data corresponding to the spatial resolution of 15 meters; the voting method that based on the ALOS and ASTER data combination is the best fusion strategy in decision level fusion.

Keywords： remote sensing coal mining collapse seeper scale effect decision fusion