为解决稀土矿开采区野外调查用时长、成本高等问题,文中利用遥感影像来提取地物信息.全文基于资源一号02C卫星影像,以赣州市寻乌县文峰乡一稀土矿区为研究区,对影像数据进行了预处理,采用最佳的Pan Sharpening融合结果进行多尺度分割,结合目标地物的光谱特征、空间特征、纹理特征等,设置分类规则进行面向对象的信息提取研究,分类结果总体精度为78.71%,达到较好的分类质量.研究结果证明:利用遥感影像提取地物信息具有用时短、成本低、信息量大的优势.%To cope with the long duration and high cost involved in the field survey in rare earth mining area, the remote-sensing image is used in this paper to extract the information on ground objects. The whole research, in line with ZY-1 02C image of a rare earth mining area in Wenfeng Township, Xunwu County, Ganzhou City, pre-processes the imaging data, fuses the 2.36m panchromatic images with 10m multispectral images in multiple fusion methods, segmentates the best Pan Sharpening fusion results, sets up the classification rules and researches the extracted object-oriented information in accordance with the spectral characteristics, spatial characteristics, texture characteristics of targeted ground objects. The overall accuracy of the classification results is 78.71%, which achieves a better classification quality. The results show that using remote sensing image to extract the information of ground objects has advantages of short time, low cost and abundant information.
展开▼
