作者: Yuehong Chen , Yong Ge , Yu Chen , Yan Jin , Ru An
DOI: 10.1109/TGRS.2018.2808410
关键词:
摘要: This paper proposes a new subpixel mapping (SPM) method based on multiscale spatial dependence (MSD). At the beginning, it adopts object-based and pixel-based soft classifications to generate class proportions within each object pixel, respectively. Then, object-scale of land cover classes is extracted from objects, combined at both pixel scale obtained pixels. Furthermore, these dependences are fused as MSD for subpixel. Last, linear optimization model built determine where spatially distribute mixed scales. Three experiments two synthetic images real remote sensing image carried out evaluate effectiveness MSD. The experimental results show that performed better than four existing SPM methods by generating less isolated classified pixels those generated three more local details an method. Hence, provides valuable solution producing maps