Direct curvature scale space in corner detection

Authors:
Baojiang Zhong;Wenhe Liao
Affiliations:
Department of Mathematics, Nanjing university of Aeronautics, & Astronautics, Nanjing, China;College of Mechanical and Electrical Engineering, Nanjing university of Aeronautics & Astronautics, Nanjing, China
Venue:
SSPR'06/SPR'06 Proceedings of the 2006 joint IAPR international conference on Structural, Syntactic, and Statistical Pattern Recognition
Year:
2006

Citing 9
Cited 0

The Curvature Primal Sketch

IEEE Transactions on Pattern Analysis and Machine Intelligence
Scale-Based Description and Recognition of Planar Curves and Two-Dimensional Shapes

IEEE Transactions on Pattern Analysis and Machine Intelligence
Scale-Based Detection of Corners of Planar Curves

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Theory of Multiscale, Curvature-Based Shape Representation for Planar Curves

IEEE Transactions on Pattern Analysis and Machine Intelligence
Curvature Scale Space Representation: Theory, Applications, and MPEG-7 Standardization

Curvature Scale Space Representation: Theory, Applications, and MPEG-7 Standardization
A Hybrid Method for Fast Computing the Curvature Scale Space Image

GMP '04 Proceedings of the Geometric Modeling and Processing 2004
Scale-space filtering

IJCAI'83 Proceedings of the Eighth international joint conference on Artificial intelligence - Volume 2
Perceptually relevant and piecewise linear matching of silhouettes

Pattern Recognition
Robust automatic selection of optimal views in multi-view free-form object recognition

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Curvature Scale Space (CSS) representation of planar curves is considered to be a modern tool in image processing and shape analysis. Direct Curvature Scale Space (DCSS) is defined as CSS that results from convolving the curvature of a curve with a Gaussian kernel directly. Recently a theory of DCSS in corner detection has been established. In the present paper the DCSS theory is considered to transform the DCSS image of a given curve into a tree organization, and then corners on the curve are detected and located in a multiscale sense. Experiments are conducted to show that the DCSS corner detector can work equally well as the CSS corner detector does on curves with multiple-size features, however, at much less computational cost.