论文部分内容阅读
针对多时相、多分辨率遥感影像数据的特点,充分考虑不同分辨率数据和不同变化检测应用的需求,将由粗到精数据集分层检测和决策级融合的思想引入到变化检测,以多时相多分辨率ALOS遥感影像为例,构建并试验了由粗到精变化检测的技术流程。该方法将ALOS多光谱数据视为粗数据集,将全色数据和融合数据视为精数据集,通过对3种数据集变化检测结果按照一定的决策规则进行综合,生成最终的变化检测结果图,反映变化的发生位置及变化强度。选择江苏省徐州市城区和矿区两个区域进行试验并与常规变化检测算法结果进行对比,表明该方法具有更好的检测效果,可以有效地应用于多分辨率遥感影像变化检测,并为实际野外调查提供重要的检测靶区。
Considering the characteristics of multi-temporal and multiresolution remote sensing image data, taking into account the needs of different resolution data and different detection applications, this paper introduces the idea of hierarchical and decision-level fusion from coarse to fine data set to change detection, Multiresolution ALOS remote sensing image is taken as an example to construct and test the technological process from coarse to fine variation detection. The method regards ALOS multispectral data as a rough dataset, panchromatic datum and fusion datum as a fine dataset, and synthesizes the results of the three dataset changes according to certain decision rules to generate the final change test result graph , Reflecting the location and intensity of change. The experiment of urban and mining area in Xuzhou City, Jiangsu Province was conducted and compared with the results of routine change detection algorithm. The results show that this method has better detection effect and can be effectively applied to the detection of multi-resolution remote sensing image change. The survey provides important test targets.