A New Color Blindness Cure Model Based on BP Neural Network
There is no method of color blindness cure so far although some research on color blindness has been done. In this paper we propose the BP neural network model for Color Blindness Cure (CBC) according to the biological experimental results. This model transforms abnormal output signals of two paths corresponding to two kinds of cones of the dichromats into normal ones of three paths corresponding to three kinds of cones of normal people. Computer simulation results show that this model has good performance and changes the abnormal vision of the dichromats into normal ones. In addition, some improvements of our model are introduced.
KeywordsNeural Network Output Signal Hide Neuron Normal People Color Blindness
Unable to display preview. Download preview PDF.
- 1.Wandell, B.A.: Foundations of Vision. Sinauer Associates Inc. Publishers, Sunderland (1995)Google Scholar
- 2.Hubel, D.H.: Eye, Brain, and Vision. Scientific American Library Series, New York (1988)Google Scholar
- 4.Nakauchi, S., Usui, S.: Multilayered Neural Network Models for Color Blindness. In: IEEE International Joint Conference on Neural Networks 1991, vol. 1, pp. 473–478 (1991)Google Scholar
- 5.Kuo, Y.H., Hsu, J.P.: MCFC-R: A Fuzzy Connectionist Model for Color-Blindness Plate Recognition. In: IEEE International Conference on Neural Networks 1996, vol. 2, pp. 718–723 (1996)Google Scholar
- 9.Faugeras, O.D.: Digital Color Image Processing and Psychophysics within the Framework of a Human Visual Model. Ph.D. dissertation, Univ. Utah, Salt Lake City (1976)Google Scholar