Adaptive Curved Feature Detection Based on Ridgelet
Feature detection always is an important problem in image processing. Ridgelet performs very well for objects with linear singularities. Based on the idea of ridgelet, this paper presents an adaptive algorithm for detecting curved feature in anisotropic images. The curve is adaptively partitioned into fragments with different length, and these fragments are nearly straight at fine scales, then it can be detected by using ridgelet transform. Experimental results prove the efficiency of this algorithm.
KeywordsAdaptive Algorithm Curve Singularity Speckle Noise Curve Feature Break Part
Unable to display preview. Download preview PDF.
- 1.Candès, E.J.: Ridgelets: Theory and Application. PhD Thesis, Department of Statistics, Stanford University (1998) Google Scholar
- 2.Candès, E.J.: Monoscale Ridgelets for the Representation of Images with Edges [R]. Department of Statistics, Stanford University (1999) Google Scholar
- 3.Averbuch, A., Coifman, R.R., Donoho, D.L., Israeli, M. and Walden, J.: Fast Slant Stack : A Notion of Radon Transform for Data in a Cartesian Grid which is Rapidly Computible, Algebraically Exact, Geometrically Faithful an Invertible. Department of Statistics, Stanford University (2001) Google Scholar
- 4.Candès, E.J. Donoho, D.L.: Recovering Edges in Ill-Posed Inverse Problems: Optimality of Curvelet Frames. Department of Statistics, Stanford University (2000) Google Scholar
- 6.Candès, E.J. and Donoho, D.L.: Ridgelts: A Key to Higher-Dimensional Intermittency? Department of Statistics, Stanford University (1999) Google Scholar