Land Remote Sensing Classification Method of Open-pit Mines Based on Three-parameters Threshold Segmentation
Thetraditional single-machine remote sensing image processing mode has low efficiency and is difficult to meet the demand of rapid classification of mine land cover in any region of the world.Based on the remote sensing cloud computing platform,this paper developed a three-parameters threshold segmentation and classification method,and selected 9 open-pit mines worldwide to complete the verification.The results show that:1)NDVI,NDWI and NDBI are sensitive indexes of mine land cover,which can effectively dis-tinguish mine land cover types.2)Based on remote sensing cloud computing platform and three-parameters threshold segmentation method,the rapid classification of land cover of mines across the world can be realized.The overall accuracy of three-parameters threshold segmentation method is higher than that of supervised classification.3)The classification results of the three-parameters threshold segmentation method have clear boundaries,and the classification accuracy order is bare soil>forest>water>bare rock>shrub.The segmentation threshold is adaptable to mines in different climatic zones of the earth.In general,the three-parameters threshold segmentation method based on remote sensing cloud computing platform can effectively realize the rapid classification of mine land cover in different regions,which can provide a new method for the global monitoring of mine land cover in the future.
万方数据
维普
