Nearest-neighbour classifiers in natural scene analysis. (English) Zbl 0984.68709

Summary: It is now well-established that \(k\) nearest-neighbour classifiers offer a quick and reliable method of data classification. In this paper we extend the basic definition of the standard \(k\) nearest-neighbour algorithm to include the ability to resolve conflicts when the highest number of nearest neighbours are found for more than one training class (model-1). We also propose model-2 of nearest-neighbour algorithm that is based on finding the nearest average distance rather than nearest maximum number of neighbours. These new models are explored using image understanding data. The models are evaluated on pattern recognition accuracy for correctly recognising image texture data of five natural classes: grass, trees, sky, river reflecting sky and river reflecting trees. On noise contaminated test data, the new nearest neighbour models show very promising results for further studies. We evaluate their performance with increasing values of neighbours \((k)\) and discuss their future in scene analysis research.


68U99 Computing methodologies and applications
68T45 Machine vision and scene understanding


scene analysis
Full Text: DOI


[1] Kodratoff, Y.; Moscatelli, S., Machine learning for object recognition and scene analysis, Int. J. Pattern Recognition Artif. Intell., 8, 1, 259-304 (1994)
[6] Firschein, O., Defence applications of image understanding, IEEE Expert, 10, 5, 11-17 (1995)
[7] Simpson, R. L., Computer visionan overview, IEEE Expert, 6, 4, 11-15 (1991)
[8] Campbell, N. W.; Mackeown, W. P.J.; Thomas, B. T.; Troscianko, T., Interpreting image databases by region classification, Pattern Recognition, 30, 4, 555-563 (1997)
[9] Liu, H.; Yun, D. Y.Y., Adaptive image segmentation by quantisation, Proc. SPIE, 1766, 322-332 (1992)
[10] Lakany, H. M.; Schukat-Talamazzini, E. G.; Niemann, H., Object recognition from 2D images using Kohonen self-organised feature maps, Pattern Recognition Image Anal., 7, 3, 301-308 (1997)
[11] Kasparis, T.; Eichmann, G.; Georgiopoulos, M.; Hieleman, G. L., Image pattern algorithms using neural networks, Proc. SPIE, 1297, 298-306 (1990)
[12] Booth, R.; Allen, C. R., A neural network implementation for real-time scene analysis, Proc. SPIE, 1001, 2, 1086-1092 (1988)
[13] Beveridge, J. R.; Griffith, J.; Kohler, R. R.; Hanson, A. R.; Riseman, E. M., Segmenting images using localised histograms and region merging, Int. J. Comput. Vision, 2, 3, 311-347 (1989)
[14] Dellepiane, S.; Vernazza, G., A fuzzy approach to cue detection and region merging for image segmentation, (Cantoni, V.; Creutzburg, R.; Levialdi, S.; Wolf, G., Recent Issues in Pattern Analysis and Recognition (1989), Springer: Springer Berlin), 58-64
[16] Sadjadi, F., Performance evaluation of a texture-based segmentation algorithm, Proc. SPIE, 1483, 185-195 (1991)
[17] Theodoridis, S.; Koutroumbas, K., Pattern Recognition (1999), Academic Press: Academic Press New York
[18] Singh, S., A single nearest neighbour fuzzy approach for pattern recognition, Int. J. Pattern Recognition Artif. Intell., 13, 1, 49-54 (1999)
[19] Haddon, J. F.; Boyce, J. F., Integrating spatio-temporal information in image sequence analysis for the enforcement of consistency of interpretation, Digital Signal Processing, 8, 4, 284-293 (1998)
[20] Haddon, J. F.; Boyce, J. F., Image segmentation by unifying region and boundary information, IEEE Trans. Pattern Anal. Mach. Intell., 12, 10, 929-948 (1990)
[21] Haralick, R. M.; Shanmugan, K.; Dinstein, I., Texture features for image classification, IEEE, SMC-3, 6, 610-621 (1973)
[22] Haralick, R. M., Image texture survey, (Krishnaiah, P. R.; Kanal, L. N. (1982), Handbook of Statistics: Handbook of Statistics Vol. 2), 399-415
[23] Haddon, J. F.; Boyce, J. F., Co-occurrence matrices for image analysis, IEE Electron. Commun. Eng. J., 5, 2, 71-83 (1993)
[27] Singh, S., Effect of noise on generalisation in massively parallel fuzzy systems, Pattern Recognition, 31, 11, 25-33 (1998)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.