Surface Curvature as a Measure of Image Texture

Authors:
F. G. Peet;T. S. Sahota
Affiliations:
Pacific Forest Research Centre, Canadian Forestry Service, Victoria, B.C., Canada V8Z 1M5.;Pacific Forest Research Centre, Canadian Forestry Service, Victoria, B.C., Canada V8Z 1M5.
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
1985

Citing 0
Cited 5

An automated facial pose estimation using surface curvature and tetrahedral structure of a nose

ACIVS'05 Proceedings of the 7th international conference on Advanced Concepts for Intelligent Vision Systems
Soft computing based range facial recognition using eigenface

ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part IV
3D facial image recognition using a nose volume and curvature based eigenface

GMP'06 Proceedings of the 4th international conference on Geometric Modeling and Processing
Continuous rotation invariant local descriptors for texton dictionary-based texture classification

Computer Vision and Image Understanding
Multi-focus image fusion based on the neighbor distance

Pattern Recognition

Quantified Score

Hi-index	0.15

Visualization

Abstract

Methods from the theory of surfaces in differential geometry are applied to the gray tone intensity and absorbance surfaces defined by a digital image. In particular, the first fundamental form, the second fundamental form, the two principal curvatures, the Gaussian curvature, and mean curvature from classical differential geometry are presented. New curvatures are introduced which appear to be more appropriate for the purposes of image texture. For each curvature, a texture feature is generated. These features are applied to biological cell nuclei and it is found that they are useful for purposes of discrimination.