On Resolving Ambiguities in Arbitrary-Shape extraction by the Hough Transform

The Hough transform extracts a shape by gathering evidence obtained by mapping points from the image space into a parameter space. In this process, wrong evidence is generated from image points that do not correspond to the model shape. In this paper, we show that significant wrong evidence can be generated when the Hough Transform is used to extract arbitrary shapes under rigid transformations. In order to reduce the wrong evidence, we consider two types of constraints. First, we define constraints by considering invariant features. Secondly, we consider constraints defined via a gradient direction information. Our results show that these constraints can significantly improve the gathering strategy, leading to identification of the correct parameters. The presented formulation is valid for any rigid transformations represented by affine mappings.

[1]  Josef Kittler,et al.  A survey of the hough transform , 1988, Comput. Vis. Graph. Image Process..

[2]  W. Eric L. Grimson,et al.  On the Sensitivity of the Hough Transform for Object Recognition , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Rangachar Kasturi,et al.  Pose clustering on constraints for object recognition , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Peter Kwong-Shun Tam,et al.  Modification of hough transform for object recognition using a 2-dimensional array , 1995, Pattern Recognit..

[5]  Mark S. Nixon,et al.  Extracting arbitrary geometric primitives represented by Fourier descriptors , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[6]  Saburo Tsuji,et al.  Detection of Ellipses by a Modified Hough Transformation , 1978, IEEE Transactions on Computers.

[7]  Wen-Hsiang Tsai,et al.  Perspective-transformation-invariant generalized hough transform for perspective planar shape detection and matching , 1997, Pattern Recognit..

[8]  Wen-Yen Wu,et al.  Elliptical object detection by using its geometric properties , 1993, Pattern Recognit..

[9]  Mark S. Nixon,et al.  On using directional information for parameter space decomposition in ellipse detection , 1996, Pattern Recognit..

[10]  Yehezkel Lamdan,et al.  On the error analysis of 'geometric hashing' , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[11]  Suchendra M. Bhandarkar,et al.  Sensitivity analysis for matching and pose computation using dihedral junctions , 1991, Pattern Recognit..

[12]  Jack Sklansky,et al.  On the Hough Technique for Curve Detection , 1978, IEEE Transactions on Computers.

[13]  Ishwar K. Sethi,et al.  An ellipse detection method from the polar and pole definition of conics , 1993, Pattern Recognition.

[14]  Rakesh Mohan,et al.  Multidimensional indexing for recognizing visual shapes , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[15]  S. Y. Yuen,et al.  An investigation of the nature of parametrization for the Hough transform , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[16]  Dana H. Ballard,et al.  Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[17]  W. Eric L. Grimson,et al.  On the Verification of Hypothesized Matches in Model-Based Recognition , 1991, IEEE Trans. Pattern Anal. Mach. Intell..