Object-based change monitoring in mining areas—taking Pingshuo as an example

In this paper, RapidEye Remote Sensing Image (Map 2012) and SPOT7 Remote Sensing Image (Map 2015) in Pingshuo Mining Area are selected as the data. Combined with object-based classification and change vector analysis method, we studied the feasibility of high resolution remote sensing image for mining land classification and the accuracy of monitoring. The results show that the classification of reclaimed mining land has higher precision, the overall accuracy and kappa coefficient of the classification of Map2012 were 0.89 and 0.87, and the change region map were 0.87 and 0.84. It’s obvious that object-based classification and change vector analysis which has a great significance to improve the monitoring accuracy can be used to monitor mining land, especially reclaiming mining land.