Abstract
This paper presents a hierarchical modular neural network for colour classification in graphic arts, capable of distinguishing among very similar colour classes. The network performs analysis in a rough to fine fashion, and is able to achieve a high average classification speed and a low classification error. In the rough stage of the analysis, clusters of highly overlapping colour classes are detected. Discrimination between such colour classes is performed in the next stage by using additional colour information from the surroundings of the pixel being classified. Committees of networks make decisions in the next stage. Outputs of members of the committees are adaptively fused through the BADD defuzzification strategy or the discrete Choquet fuzzy integral. The structure of the network is automatically established during the training process. Experimental investigations show the capability of the network to distinguish among very similar colour classes that can occur in multicoloured printed pictures. The classification accuracy obtained is sufficient for the network to be used for inspecting the quality of multicoloured prints.
Similar content being viewed by others
References
Malmqvist K, Malmqvist L, Verikas A, Bergman L. A new method for colour measurements in multicolour newspaper pictures and its use for ink feed control. Proc Int Conf ‘Printing and Graphic Arts’, Mineapolis, USA, 1996; 181–185
Liu J, Yang YH. Multiresolution color image segmentation. IEEE Trans Pattern Analysis Machine Intelligence 1994; 16(7): 689–700
Panjwani DK, Healey G. Markov Random Field models for unsupervised segmentation of textured color images. IEEE Trans Pattern Analysis Machine Intelligence 1995; 17(10): 939–954
Tominaga S. Color classification of natural color images. Color Res Applic 1992; 17(4): 230–239
Uchiyama M, Arbib MA. Color image segmentation using competitive learning. IEEE Trans Pattern Analysis Machine Intelligence 1994; 16(12): 1197–1206
Verikas A, Malmqvist K, Bergman L. Colour image segmentation by modular neural network. Pattern Recognition Lett 1997; 18(2): 173–185
Verikas A, Malmqvist K, Bergman L, Gelzinis A. An unsupervised learning technique for finding decision boundaries. Proc 5th Euro Conf Artificial Neural Networks (ICANN-95), Paris, 1995; 2: 99–104
Verikas A, Malmqvist K, Gelzinis A. A new technique to generate a binary decision tree. Proc Symposium Image Analysis, Lund, Sweden, 1996; 164–168
Jacobs RA, Jordan MI, Nowlan SJ, Hinton GE. Adaptive mixture of local experts. Neural Computation 1991; 3: 79–87
Jordan MI, Jacobs R. Hierarchical mixture of experts and the EM algorithm. Neural Computation 1994; 6: 79–87
Xu L, Krzyzak A, Suen CY. Methods for combining multiple classifiers and their applications to handwriting recognition. IEEE Trans Systems, Man Cybernetics 1992; 22(3): 418–435
Lam L, Suen CY. Optimal combination of pattern classifiers. Pattern Recognition Lett 1995; 16: 945–954
Woods K, Kegelmeyer WP, Bowyer K. Combination of multiple classifiers using local accuracy estimates. IEEE Trans Pattern Analysis Machine Intelligence 1997; 19(4): 405–410
Ho TK, Hull JJ, Srihari SN. Decision combination in multiple classifier systems. IEEE Trans Pattern Analysis Machine Intelligence 1994; 16(1): 66–75
Rogova G. Combining the results of several neural network classifiers. Neural Networks 1994; 7(5): 777–781
Cho SB, Kim JH. Combining multiple neural networks by fuzzy integral for robust classification. IEEE Trans Systems, Man Cybernetics 1995; 25(2): 380–384
Gadner PD, Mohamed MA, Keller JM. Fusion of handwritten word classifiers. Pattern Recognition Lett 1996; 17: 577–584
Sollich P, Krogh A. learning with ensembles: How over-fitting can be useful. In Advances in Neural Information Processing Systems 8, Touretzky DS, Mozer MC, Hasselmo ME. (eds.), MIT Press, 1996; 190–197
Krogh A, Vedelsby J. Neural network ensembles, cross validation, and active learning. In Advances in Neural Information Processing Systems 7, Tesauro G, Touretzky DS, Leen TK. (eds.), MIT Press, 1995; 231–238
Tresp V, Taniguchi M. Combining estimators using non-constant weighting functions. In Advances in Neural Information Processing Systems 7, Tesauro G, Touretzky DS, Leen TK. (eds.), MIT Press, 1995
Sugeno M. Fuzzy measures and fuzzy integrals: a survey. In Fuzzy Automata and Decision Processes, Amsterdam: North Holland, 1977; 89–102
Verikas A, Signahl M, Malmqvist K, Bacauskiene M. Fuzzy committee of experts for segmentation of colour images. Proc 5th Euro Congress on Intelligent Techniques and Soft Computing, Aachen, Germany, 1997; 3: 1902–1906
Verikas A, Malmqvist K, Malmaqvist L, Bergman L. A new method for colour measurements in graphic arts. Colour Res Applic 1998 (submitted)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Verikas, A., Malmqvist, K., Bergman, L. et al. Colour classification by neural networks in graphic arts. Neural Comput & Applic 7, 52–64 (1998). https://doi.org/10.1007/BF01413709
Issue Date:
DOI: https://doi.org/10.1007/BF01413709