Image segmentation using graph cut technique for outdoor scene images

This paper proposes an outdoor scene image segmentation using graph cut method based on their color information. The basic aim is to differentiate the background (such as sky, ground, vegetation etc) and foreground (such as buildings people etc) of the outdoor scene images based on their color and texture information. This paper shows the comparison of results obtained using graph cut method, graph cut based on color information and also with other techniques of image segmentation such as Clustering techniques and threshold techniques. Graph cut is a popular method for interactive image segmentation. A graph cut is the process of partitioning a directed or undirected graph into disjoint sets. The results obtained by this graph cut with color information are accurate and satisfactory which highlights the foreground and darkens the background.