Multi-feature combined cloud and cloud shadow detection in GaoFen-1 wide field of view imagery

摘要

The wide field of view (WFV) imaging system onboard the Chinese GaoFen-1(GF-1) optical satellite has a 16-m resolution and four-day revisit cycle forlarge-scale Earth observation. The advantages of the high temporal-spatialresolution and the wide field of view make the GF-1 WFV imagery very popular.However, cloud cover is an inevitable problem in GF-1 WFV imagery, whichinfluences its precise application. Accurate cloud and cloud shadow detectionin GF-1 WFV imagery is quite difficult due to the fact that there are onlythree visible bands and one near-infrared band. In this paper, an automaticmulti-feature combined (MFC) method is proposed for cloud and cloud shadowdetection in GF-1 WFV imagery. The MFC algorithm first implements thresholdsegmentation based on the spectral features and mask refinement based on guidedfiltering to generate a preliminary cloud mask. The geometric features are thenused in combination with the texture features to improve the cloud detectionresults and produce the final cloud mask. Finally, the cloud shadow mask can beacquired by means of the cloud and shadow matching and follow-up correctionprocess. The method was validated using 108 globally distributed scenes. Theresults indicate that MFC performs well under most conditions, and the averageoverall accuracy of MFC cloud detection is as high as 96.8%. In the contrastiveanalysis with the official provided cloud fractions, MFC shows a significantimprovement in cloud fraction estimation, and achieves a high accuracy for thecloud and cloud shadow detection in the GF-1 WFV imagery with fewer spectralbands. The proposed method could be used as a preprocessing step in the futureto monitor land-cover change, and it could also be easily extended to otheroptical satellite imagery which has a similar spectral setting.