Abstract
Edge linking is a fundamental computer vision task, yet presents difficulties arising from the lack of information in the image. Viewed as a constrained optimisation problem, it is NP hard — being isomorphic to the classical travelling salesman problem. Self-learning neural techniques boast the ability to solve hard, ill-defined problems, and hence offer promise for such an application. This paper examines the suitability of four well-known unsupervised techniques for the task of edge linking, by applying them to a test bed of edge point images and then evaluating their performance both quantitatively and qualitatively. Techniques studied are the elastic net, active contours, Kohonen map and Burr's modified elastic net. Of these, only the elastic net and the Kohonen map are realistic contenders for general edge-linking tasks. However, the other two exhibit behaviour which may make them particularly suited to some specific image-processing and computer vision applications.
Similar content being viewed by others
References
Hertz J, Krogh A, Palmer RG. Introduction to the Theory of Neural Computation. Addison-Wesley, 1991.
Hopfield JJ, Tank DW. Neural computation of decisions in optimization problems. Biol Cybern 1985; 52: 141–152.
Wilson GV, Pawley GS. On the stability of the travelling salesman problem algorithm of Hopfield and Tank. Biol Cybern 1988; 58: 63–70.
Durbin R, Willshaw D. An analogue approach to the travelling salesman problem using an elastic net method. Nature 1987; 326: 689–691.
Simic PD. Statistical mechanics as the underlying theory of ‘elastic’ and ‘neural’ optimisations. Network 1990; 1: 89–103.
Kirkpatrick S, Gelatt CD, Vecchi MP. Optimization by simulated annealing. Science 1983; 220: 671–680.
Abrantes AJ, Marques JS. Unified approach to snakes, elastic nets, and Kohonen maps. Proc IEEE Int Conf Acoustics, Speech and Signal Processing (ICASSP '95) May 1995, vol. 5: 3427–3430.
Kohonen T. Clustering, taxonomy, and topological maps of patterns. Proc 6th Int Conf Pattern Recognition, October 1982; 114–128.
Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models. Int J Computer Vision 1987; 1: 321–331.
Burr DJ. An improved elastic net method for the TSP. IEEE Int Conf Neural Networks 1988, vol 1; 69–76.
Xu G, Segawa E, Tsuji S. Robust active contours with insensitive parameters. Pattern Recognition 1994; 27(7): 879–884.
Kohonen T. The ‘neural’ phonetic typewriter. IEEE Computer Magazine 1988; 11–22.
Ahalt SC, Krishnamurthy AK, Chen P, Melton DE. Competitive learning algorithms for vector quantisation. Neural Networks 1990; 3: 277–290.
Nasrabadi NM, Feng Y. Quantisation of images based upon the Kohonen self organising feature maps. IEEE Int Conf Neural Networks, 1988; 101–108.
Williams C. Combining Deformable Models and Neural Networks for Hand-written Digit Recognition. PhD thesis, University of Toronto, November 1994.
Amin S. A self-organised travelling salesman. Neural Comput & Applic 1994; 2: 129–133.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Gilson, S.J., Damper, R.I. An empirical comparison of neural techniques for edge linking of images. Neural Comput & Applic 6, 64–78 (1997). https://doi.org/10.1007/BF01414004
Issue Date:
DOI: https://doi.org/10.1007/BF01414004