Title :
Similarity measure of labelled images
Author :
Horiuchi, Takahiko
Author_Institution :
Dept. of Inf. & Image Sci., Chiba Univ., Japan
Abstract :
A similarity measure is developed for comparing labelled images. The proposed measure is calculated based on binary relations of arbitrary pixels in the labelled images. Hence, the proposed measure can evaluate the difference between two images speedy by pixel-to-pixel calculation. The measure can be used for not only spatial labelled images by area-based labelling such as region segmentation, but also colour labelled images by pixel-based labelling such as colour clustering. A few actual applications are shown in order to explain the behaviour.
Keywords :
image colour analysis; image segmentation; pattern clustering; area based labelling; colour clustering; colour labelled images; image segmentation; image similarity measure; pixel based labelling; Area measurement; Color; Digital images; Image segmentation; Labeling; Pixel; Quantization; Shape measurement; Stability; Velocity measurement;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334601