Premium
Detection and Removal of Clouds and Associated Shadows in Satellite Imagery Based on Simulated Radiance Fields
Author(s) -
Wang Tianxing,
Shi Jiancheng,
Letu Husi,
Ma Ya,
Li Xingcai,
Zheng Yaomin
Publication year - 2019
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1029/2018jd029960
Subject(s) - radiance , remote sensing , moderate resolution imaging spectroradiometer , satellite , preprocessor , computer science , shadow (psychology) , spectroradiometer , cloud computing , environmental science , computer vision , artificial intelligence , geology , reflectivity , physics , optics , psychology , astronomy , psychotherapist , operating system
Clouds and shadows pose a significant barrier for land surface optical and infrared remote sensing image processing and their various applications. The detection and removal of clouds and shadows from satellite images have always been critical preprocessing steps. To date, a variety of methods have been designed to solve this problem. Some require particular channels, while others are heavily dependent on the availability of temporally adjacent images (reference images). Moreover, many methods are too complex to use by common users. For those reasons, in this paper an alternative scheme for detecting clouds and shadows is proposed based on simulated top‐of‐atmosphere radiance fields. At the same time, a simple approach to remove clouds and shadows is also provided. The results indicate that the new method can properly identify both clouds and shadows in satellite images. Especially, it shows obvious advantage over the Moderate Resolution Imaging Spectroradiometer cloud product (MOD35) for shadow detection. Although the proposed cloud removal method is simple, the radiances of a contaminated image can be reasonably reconstructed with root‐mean‐square error < 3.0 W/m 2 ·sr·μm and mean bias < 1.0 W/m 2 ·sr·μm for all seven Moderate Resolution Imaging Spectroradiometer reflective bands for our case studies. These results prove the effectiveness of the proposed scheme in identifying and removing clouds and shadows from remotely sensed images. Meanwhile, these findings provide some new ideas for the remote sensing community, especially in the fields of cloud detection and image processing.