基于子像元交换算法和地形数据的土地覆盖亚像元制图研究

发布时间：2018-07-21 16:27

【摘要】：在将遥感影像应用于土地覆盖制图的过程中,混合像元会产生负面影响,尤其是在空间分辨率较低的情况下。软分类和超分辨率(亚像元/子像元)制图技术可以解决上述问题。子像元交换算法是一种简单而有效的亚像元制图技术,但也存在计算效率不够高和在超分辨率制图因子(scale factor,S)较大时制图精度较差的问题,其原因可能是亚像元制图只是从软分类结果中获得信息。因此,将数字高程模型(digital elevation model,DEM)及其衍生的数据作为子像元交换算法的辅助信息,研究提高其制图精度的有效性。研究结果表明:(1)在DEM信息的辅助下,亚像元制图的精度得到了改善,即使在S较大时也是有效的;(2)同时使用高程和坡度信息时的效果要好于单独应用高程或坡度信息;(3)在S较大的情况下,制图精度对邻域范围大小的敏感性要低于S较小时;(4)在DEM信息的辅助下,计算效率得到提高。实验结果表明,将DEM作为辅助信息进行亚像元制图是有效和可行的。
[Abstract]:In the process of applying remote sensing images to land cover mapping, mixed pixels will have a negative effect, especially in the case of low spatial resolution. Soft classification and super-resolution (sub-pixel / sub-pixel) mapping techniques can solve these problems. Subpixel exchange algorithm is a simple and effective sub-pixel mapping technique, but it also has some problems such as low computational efficiency and poor mapping accuracy when the scale factor S is larger. The reason may be that sub-pixel mapping only obtains information from soft classification results. Therefore, the digital elevation model (digital elevation model Dem) and its derived data are taken as the auxiliary information of the sub-pixel exchange algorithm to study the effectiveness of improving the mapping accuracy of the digital elevation model (digital elevation model / Dem). The results show that: (1) with the help of Dem information, the accuracy of sub-pixel mapping is improved. It is effective even when S is larger; (2) it is better to use elevation and slope information at the same time than using elevation or gradient information alone; (3) when S is larger, The sensitivity of mapping precision to neighborhood range is lower than that of S, and (4) with the help of Dem information, the computational efficiency is improved. The experimental results show that it is effective and feasible to use Dem as auxiliary information for sub-pixel mapping.
【作者单位】：浙江大学环境与资源学院;杭州学联土地规划设计咨询有限公司;
【基金】：国家科技支撑计划项目“旱区多遥感平台农田信息精准获取技术集成与服务”(编号:2012BAH29B04)资助
【分类号】：F301;TP751

【相似文献】