F-1333 Image Segmentation Using Multi Dimensional Co-occurrence Matrix (2nd Report)
Yoshihiko Nomura, Terumoto Komori, Jun Ohmura, James F Boyce · The proceedings of the JSME annual meeting · 2001
The co-occurrence matrix based method of image segmentation belongs to the class of region-based techniques. It utilizes the observations: (1) features of the multiple windows neighboring a pixel do not differ significantly from one another, (2) features corresponding to pixels belonging to the same object form a cluster in the feature space, (3) the cluster may frequently be approximated by a Gaussian distribution. This paper extends the co-occurrence matrix based method. The definition of co-occurrence features is extended from one dimension to many dimensions: the number of observation windows is extended from two to an arbitrary number.