Abstract
Extraction of the skeletal shape of an elongated object is often required in object recognition and classification problems. Various techniques have so far been developed for this purpose. A comprehensive comparative study is carried out here between neural network-based and conventional techniques. The main problems with the conventional methods are noise sensitivity and rotation dependency. Most of the existing algorithms are sensitive to boundary noise and interior noise. Also, they are mostly rotation dependent, particularly if the angle of rotation is not a multiple of 90°. On the other hand, the neural network based technique discussed here is found to be highly robust in terms of boundary noise as well as interior noise. The neural method produces satisfactory results even for a very low (close to 1) Signal to Noise Ratio (SNR). The algorithm is also found to be efficient in terms of invariance under arbitrary rotations and data reduction. Moreover, unlike the conventional algorithms, it is grid independent. Finally, the neural technique is easily extendible to dot patterns and grey-level patterns also.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Smith RW. Computer processing of line images: a survey. Pattern Recognition 1987; 20: 7–15
Lam L, Lee SW, Suen CY. Thinning methodologies — a comprehensive survey. IEEE Trans. PAMI 1992; 14: 869–885
Datta A, Pal T, Parui SK. A modified self-organizing neural net for shape extraction. Neurocomputing 1997; 14: 3–14
Kohonen T. Self-Organization and Associative Memory, Springer-Verlag, 1989
Sabourin M, Mitiche A. Modeling and classification of shape using a Kohonen associative memory with selective multiresolution. Neural Networks 1993; 6: 275–283
Fritzke B. Let it grow — self-organizing feature maps with problem dependent cell structure. In: Kohonen T et al. (eds), Artificial Neural Networks, vol. 1, North-Holland, 1991
Choi D, Park S. Self-creating and organizing neural networks. IEEE Neural Networks 1994; 5: 561–575
Kangas JA, Kohonen T, Laaksonen J. Variants of self-organizing maps. IEEE Trans Neural Networks 1990; 1: 93–99
Datta A, Parui SK. A robust parallel thinning algorithm for binary images. Pattern Recognition 1994; 27: 1181–1192
Jang BK, Chin RT. One-pass parallel thinning: analysis, properties and quantitative evaluation. IEEE Trans PAMI 1992; 14: 1129–1140
Arcelli C, Cordella L, Levialdi S. Parallel thinning of binary pictures. Electron Lett 1975; 11: 148–149
Tamura H. A comparison of line thinning algorithms from digital geometry viewpoint. Proc 4th Int Joint Conf Pattern Recog, Kyoto, Japan, 1978; 715–719
Hilditch CJ. Comparison of thinning algorithms on a parallel processor. Image Vision Comput 1983; 1: 115–132
Jang B, Chin RT. Analysis of thinning algorithms using mathematical morphology. IEEE Trans PAMI 1990; 12: 541–551
Datta A, Parui SK. Skeletons from dot patterns: a neural network approach. Pattern Recognition Letters 1997; 18: 335–342
Arcelli C, Ramella G. Finding grey-skeletons by iterated pixel removal. Image and Vision Comput 1995; 13: 159–167
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Datta, A., Parui, S.K. Shape extraction: A comparative study between neural network-based and conventional techniques. Neural Comput & Applic 7, 343–355 (1998). https://doi.org/10.1007/BF01428125
Issue Date:
DOI: https://doi.org/10.1007/BF01428125