Committee Machine for Road-Signs Classification
The paper presents a system for the road signs recognition which is based on an ensemble of the non Euclidean distance neural networks and an arbitration unit. The input to this system constitutes a binary pictogram of a sign which is supplied from the detection module. The classifier is composed of a mixture of experts – the Hamming neural networks – each working with a single group of deformed reference pictograms. The ensemble of experts is controlled by an arbitration module operating in the winner-takes-all mode. Additionally it is equipped with a promoting mechanism that favours the most populated group of unanimous experts. The presented classifier is characterized by the fast training and very fast response times which features make it suitable for the driving assistant systems. The presented concepts have been verified experimentally. Their results and conclusions are also discussed.
KeywordsInput Pattern Road Sign Driving Assistant System Winning Neuron Binary Distance
Unable to display preview. Download preview PDF.
- 1.Aoyagi, Y., Asakura, T.: A study on traffic sign recognition in scene image using genetic algorithms and neural networks. In: IEEE Conf. Electronics, pp. 1838–1843 (1996)Google Scholar
- 4.Duch, W., Grudziński, K.: A framework for similarity-based methods. In: Second Polish Conference on Theory and Applications of Artificial Intelligence, pp. 33–60 (1998)Google Scholar
- 7.Floréen, P.: Computational Complexity Problems in Neural Associative Memories. PhD Thesis, University of Helsinki, Finland (1992)Google Scholar
- 8.Lippman, R.: An introduction to computing with neural nets. IEEE Transactions on Acoustic, Speech, and Signal Processing ASSP-4, 4–22 (1987)Google Scholar
- 9.Luo, R., Potlapalli, H., Hislop, D.: Neural network based landmark recognition for robot navigation. In: IEEE Int. Conf. Industrial Electronics pp. 1084–1088 (1992)Google Scholar
- 10.Luo, R., Potlapalli, H.: Landmark recognition using projection learning for mobile robot navigation. In: Proc. IEEE Int. Conf. Neural Networks, pp. 2703–2708 (1994)Google Scholar
- 12.Road Signs and Signalization. Directive of the Polish Ministry of Infrastructure. Internal Affairs and Administration (Dz. U. Nr 170, poz. 1393) (2002)Google Scholar
- 13.Rehrmann, V., Lakmann, R., Priese, A.: A parallel system for real-time traffic sign recognition. Deimler-Benz Technical Report (1995)Google Scholar
- 14.Viola, P., Jones, M.: Rapid Object Detection using a Boosted Cascade of Simple Features. In: IEEE Conf. on Computer Vision and Pattern Recognition, pp. 151-158 (2001)Google Scholar
- 15.Zheng, Y.J., Ritter, W., Janssen, R.: An adaptive system for traffic sign recognition. In: Proc. IEEE Intelligent Vehicles Symp., pp. 165–170 (1994)Google Scholar