Description

Recognition of naturally occurring objects is a challenging task. In particular, the recognition of clouds is particularly challenging as the texture of such objects is extremely variable under different atmospheric conditions. There are several benefits of a practical system that can detect and recognize clouds in natural images especially for applications such as air traffic control. The crowning objective of this research was to identify a better cloud classification method to upgrade the current window-based clustering algorithm used operationally for China’s first operational geostationary meteorological satellite FengYun-2C (FY-2C) data. The detection of cloud in the images will be helpful for the early identification of the weather and for climate research purposes. Cloud detection using images is done based on pixel based approaches. The proposed approach uses automatic graph cut segmentation process. Otsu segmentation process is used initially employed for the identification of the cloud regions. The segmented cloud regions were classified into normal or cloudy by counting the number of segmented pixels. If the number of pixels is increased then the regions is cloudy or else the region is clear. The performance of the process is measured based on the performance metrics like Accuracy.