ABSTRACT
Sea ice concentration is an important climatic and oceanic parameter that plays an important role in the studies of heat and moisture flux between ocean and atmosphere. Besides, ice concentration has also been identified as one of the most influential factors on the magnitude of the loads on marine structures operating in the ice-covered Arctic (Antarctic) regions. The distinction of the ice pixels from water pixels is thus crucial in order to obtain the ice concentration of an ice image. The chapter introduces two simple and popular image segmentation techniques, the Otsu thresholding method and k-means clustering method. Both methods have been applied in the detection of the ice regions of various sea ice images. The thresholding method is based on the pixel's grayscale value. It extracts the objects from the background and converts the grayscale image into a binary image.
