A cloud detection algorithm-generating method for remote sensing data at visible to short-wave infrared wavelengths

• DOI: 10.1016/j.isprsjprs.2016.12.005
• Publication Date: 2016-12-31T01:31:57Z
• Authors: Lin Sun; Xueting Mi; Jing Wei; Jian Wang; Xinpeng Tian; Huiyong Yu; Ping Gan
• Keywords: Imaging spectrometer; Moderate-resolution imaging spectroradiometer; Spectral bands; Spectroradiometer

To realize highly precise and automatic cloud detection from multi-sensors, this paper proposes a algorithm-generating (CDAG) method for remote sensing data visible to short-wave infrared (SWIR) bands. Hyperspectral with high spatial resolution were collected used as pixel dataset of cloudy clear skies. In paper, multi-temporal AVIRIS (Airborne Visible/Infrared Imaging Spectrometer) 224 bands at SWIR wavelengths 20 m the dataset. Based on dataset, pixels different types clouds land cover distinguished artificially simulation multispectral sensors. Cloud algorithms sensors then generated based spectral differences between clear-sky previously. The possibility assigning was calculated reliability each method. Landsat 8 OLI (Operational Land Imager), MODIS (Moderate Resolution Spectroradiometer) Terra Suomi NPP VIIRS (Visible/Infrared Radiometer) test CDAG method, results sensor compared corresponding artificial results, demonstrating an accurate rate more than 85%.