Object-Based Superresolution Land-Cover Mapping From Remotely Sensed Imagery

Yuehong Chen, Yong Ge, Gerard B. M. Heuvelink, Ru An, Yu Chen
2018 IEEE Transactions on Geoscience and Remote Sensing  
Super resolution mapping (SRM) is a widely used technique to address mixed pixel problem in pixel-based classification. Advanced object-based classification will face the similar mixed phenomenon-mixed object that contains different land-cover classes. Currently, most SRM approaches focus on handling mixed pixels in pixel-based classification. Little if any consideration has been given to predict where classes spatially distribute within mixed objects. This article, therefore, proposes a new
more » ... ect-based super resolution mapping strategy (OSRM) to deal with mixed objects in object-based classification. First, it uses the deconvolution technique to estimate the semivariograms at target subpixel scale from the class proportions of irregular objects. Then, area-to-point kriging method is applied to predict the soft class values of subpixels within each irregular object according to the estimated semivariograms and the class proportions of objects. Finally, a linear optimization model at object-level is built to determine the optimal class labels of subpixels. Two synthetic images and a real remote sensing image were used to evaluate the performance of OSRM. The experimental results demonstrated that OSRM performed better and generated more land-cover details within mixed objects than the traditional object-based hard classification. Hence, OSRM provides a valuable solution to mixed objects in object-based classification.
doi:10.1109/tgrs.2017.2747624 fatcat:bboz24z3fbdw3eb5ynmtz3nfse