ABSTRACT
This chapter utilizes an ice shape enhancement algorithm. This can be used after the segmentation of the floes to preserve the ice floe shapes and accomplish the identification of individual ice floes and brash ice pieces. Morphological cleaning is a combination of first morphological closing and then morphological opening on an image. Both binary closing and opening operations can smooth the contours of objects, yielding results that are similar to the original shapes of the objects but with different level of details. A connected component is a set of pixels in an image that are all connected to each other. The extraction and labeling of connected components from a binary image plays a central role to many automated image processing applications. The dilation-based binary morphological reconstruction extracts the connected components of the mask image "marked" by the marker image that is contained in the mask image. It can thus be used to develop a fully automatic whole filling algorithm.
