论文部分内容阅读
采用2014年1月Quickbird拍摄的高分辨率遥感影像数据,对尾矿及固体废弃物的光谱、纹理以及空间几何特征进行分析,确定了尾矿及固体废弃物的最佳分割阈值为30和最佳合并阈值为90。通过对分类结果计算混淆矩阵,得出本次实验规则分类的总体精度为94.15%,提取精度相对较高。
Based on the high-resolution remote sensing image data collected by Quickbird in January 2014, the spectral, texture and spatial geometric characteristics of tailings and solid wastes were analyzed. The optimal thresholds of tailings and solid wastes were 30 and The best combination threshold is 90. By calculating the confusion matrix for the classification results, the overall accuracy of this experimental rule classification is 94.15%, and the extraction accuracy is relatively high.