UKIS-CSMASK: A PYTHON PACKAGE FOR MULTI-SENSOR CLOUD AND CLOUD SHADOW SEGMENTATION

Abstract. Cloud and cloud shadow segmentation is a crucial pre-processing step for any application that uses multi-spectral satellite images. In particular, time-critical disaster applications, require accurate immediate masks while being able to adapt possibly large variations caused by different sensor characteristics, scene properties or atmospheric conditions. This study introduces the newly developed open-source Python package ukis-csmask in Segmentation with performed pre-trained Convolutional Neural Network based on U-Net architecture. It works directly Level-1C data, eliminating need prior correction. Images be top of atmosphere reflectance include at least Blue, Green, Red, NIR, SWIR1 SWIR2 spectral bands. We provide performance evaluation recent benchmark dataset proof generalization ability our method across multiple satellites (Landsat-5, Landsat-7, Landsat-8, Landsat-9 Sentinel-2). also show influence augmentation image bands compare it widely used Fmask algorithm Random Forest classifier. Compared previous work this direction, focuses multi-sensor ability, simplicity efficiency provides ready-to-use software has been thoroughly tested.