CLOUD DETECTION IN SATELLITE IMAGERY USING GRAPHICS PROCESSING UNITS

Cloud detection and removal forms an important need for change detection studies. A small amount of cloud cover may misinterpret the crucial information in disaster management applications. Although several cloud detection techniques exist, there is a critical need to apply these techniques in real time and obtain the cloud free images quickly to support real time decisions. The main focus of this paper is cloud detection in multi spectral imagery using high performance technology such as graphics processing units (GPUs), and performance evaluation in terms of speed of different cloud detection techniques on GPU and CPU. The Compute Unified Device Architecture (CUDA), which is a parallel programming mechanism for GPU, is used for the implementation. Cloud detection techniques implemented are band-ratio technique, Morphological opening/closing, Morphological attribute opening/closing. Experimental results show that the speedup for GPU based implementation is around 20X for band-ratio technique, 14X for Morphological operations.