A Novel Approach for Cloud Detection in Scenes with Snow/Ice Using High Resolution Sentinel-2 Images

Ling Han,Qing Liu,Zhiheng Liu,Tingting Wu

doi:10.3390/atmos10020044

Ling Han, Qing Liu + Show 2 more

Open Access

https://doi.org/10.3390/atmos10020044

Copy DOI

Abstract

The recognition of snow versus clouds causes difficulties in cloud detection because of the similarity between cloud and snow spectral characteristics in the visible wavelength range. This paper presents a novel approach to distinguish clouds from snow to improve the accuracy of cloud detection and allow an efficient use of satellite images. Firstly, we selected thick and thin clouds from high resolution Sentinel-2 images and applied a matched filter. Secondly, the fractal digital number-frequency (DN-N) algorithm was applied to detect clouds associated with anomalies. Thirdly, spatial analyses, particularly spatial overlaying and hotspot analyses, were conducted to eliminate false anomalies. The results indicate that the method is effective for detecting clouds with various cloud covers over different areas. The resulting cloud detection effect possesses specific advantages compared to classic methods, especially for satellite images of snow and brightly colored ground objects with spectral characteristics similar to those of clouds.

Highlights

Clouds and snow cover are spectrally distinguishable despite having similar reflectance spectra in the visible-light range
The digital number-frequency (DN-N) fractal model produced false anomalies when it was used to process the curve of the Matched Filtering (MF) image generated by region of interest (ROI) 1 and ROI 2, the detected features mainly had spectral characteristics similar to those of clouds, such as snow or desert regions, indicating that the pixel value alone cannot identify cloud regions
MF image generated by ROI 1 and ROI 2, the detected features mainly had spectral characteristics to demonstrate the versatility of this algorithm

Summary

Introduction

Clouds and snow cover are spectrally distinguishable despite having similar reflectance spectra in the visible-light range. Bunting et al discussed the reflectance properties of snow and clouds in the visible and near-infrared wavelength regions [1] They analyzed imagery in pairs, with one set in the visible spectrum and the other in the near-infrared spectrum, and presented a method that utilized data from the near-infrared spectrum to analyze clouds over snow- or ice-covered regions. At these wavelengths, snow appears relatively dark while clouds are highly reflective. Cloud cover in some Antarctic images, especially on the Antarctic Peninsula, leads to the classification of clouds as rock exposure by the NDSI technique, as the two are indiscernible using this methodology [6]

Methods

Results

Conclusion