Zhanpeng Wang , Yinghai Ke, Demin Zhou, Xiaojuan Li, Lin Zhu, Huili Gong

The inevitable thick cloud contamination in Landsat images has severely limited the usability and applications of these images. Developing cloud removal algorithms has been a hot research topic in recent years. Many previous algorithms used one or multiple cloud-free images in the same area acquired on other dates as reference image(s) to reconstruct missing pixel values. Although it has been widely recognized that reference image(s) has great impacts on the performances of cloud removal algorithms, it still remains challenging to determine the optimal reference image(s). In addition, abrupt land cover change can substantially degrade the reconstruction accuracies. To address these issues, we present a new cloud removal algorithm called Virtual Image patch-based Cloud Removal (VICR). For each cloud region, VICR reconstructs the missing surface reflectance by three steps: virtual image patch construction based on time-series reference images, similar pixel selection using the newly proposed temporally weighted spectral distance (TWSD), and residual image estimation. By establishing two buffer zones around the cloud region, VICR allows automatic selection of the optimal set of time-series reference images. The effectiveness of VICR was validated at four testing sites with different landscapes (i.e., urban, croplands and wetlands) and land change patterns (i.e., phenological change, abrupt change cause by flooding and tidal inundation), and the performances were compared with mNSPI (modified neighborhood similar pixel interpolator), WLR (weighted linear regression) and ARRC (AutoRegression to Remove Clouds). Experimental results showed that VICR outperformed the other algorithms and achieved lower Root Mean Square Errors in surface reflectance estimation at the four sites. The improvement is particularly noticeable at the sites with abrupt land change. By considering the difference in the contributions from each reference image, TWSD improved the ability of VICR in predicting abrupt change in surface reflectance. Moreover, VICR is more robust to different cloud sizes and to changing reference images. VICR is also computationally faster than ARRC and mNSPI. The framework for time-series image cloud removal by VICR has great potential to be applied for large datasets processing.