TY - GEN
T1 - A new framework for hierarchical segmentation using similarity analysis
AU - Bajcsy, Peter
AU - Ahuja, Narendra
N1 - Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1997.
PY - 1997
Y1 - 1997
N2 - We present a new framework for hierarchical segmentation of multidimensional multivariate functions into homogeneous regions. Homogeneity is defined as constancy of n-th order derivatives (called features) of the function. The degree of similarity (measure of homogeneity) is used as a scale parameter to obtain a stack of segmentations. Hierarchical segmentation is represented as a tree which contains the geometric and topological information about the detected regions. Detected regions preserving their information in the tree over large range of scales are selected into a pyramid representation. Results showing noise robustness and computational efficiency of the proposed method are presented. Experiments to compare the method with three other segmentation techniques and applications to two- and three-dimensional images having one-, three- and six-variate data are described for the zeroth and first order region features.
AB - We present a new framework for hierarchical segmentation of multidimensional multivariate functions into homogeneous regions. Homogeneity is defined as constancy of n-th order derivatives (called features) of the function. The degree of similarity (measure of homogeneity) is used as a scale parameter to obtain a stack of segmentations. Hierarchical segmentation is represented as a tree which contains the geometric and topological information about the detected regions. Detected regions preserving their information in the tree over large range of scales are selected into a pyramid representation. Results showing noise robustness and computational efficiency of the proposed method are presented. Experiments to compare the method with three other segmentation techniques and applications to two- and three-dimensional images having one-, three- and six-variate data are described for the zeroth and first order region features.
UR - http://www.scopus.com/inward/record.url?scp=84957712507&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84957712507&partnerID=8YFLogxK
U2 - 10.1007/3-540-63167-4_26
DO - 10.1007/3-540-63167-4_26
M3 - Conference contribution
AN - SCOPUS:84957712507
SN - 3540631674
SN - 9783540631675
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 319
EP - 322
BT - Scale-Space Theory in Computer Vision - 1st International Conference, Scale-Space 1997, Proceedings
A2 - ter Haar Romeny, Bart
A2 - Viergever, Max
A2 - Florack, Luc
A2 - Koenderink, Jan
PB - Springer
T2 - 1st International Conference on Scale-Space Theory in Computer Vision, Scale-Space 1997
Y2 - 2 July 1997 through 4 July 1997
ER -