Title :
MCFC-R: a fuzzy connectionist model for color-blindness plate recognition
Author :
Kuo, Yau-Hwang ; Hsu, Jang-Pong
Author_Institution :
Inst. of Inf. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan
Abstract :
A fuzzy connectionist model, called MCFC-R, is developed for the pattern recognition of color images. Recognition of color-blindness plates is adopted as experimental target. This model consists of a segmentation subnetwork, a recognition subnetwork, and a reconstruction process between the two subnetworks. The two subnetworks are derived from a modified connectionist fuzzy classifier model, called MCFC. Some experiments and analyses are also made in this paper. The experimental results show that the MCFC-R model has good performance on the recognition of color-blindness plates, even in a shift-variant, scale-variant, or little data-lost situation
Keywords :
colour vision; fuzzy neural nets; image colour analysis; image recognition; image reconstruction; image segmentation; neurophysiology; physiological models; vision defects; MCFC-R; color image pattern recognition; color-blindness plate recognition; fuzzy connectionist model; little data-lost situation; recognition subnetwork; reconstruction process; scale-variant situation; segmentation subnetwork; shift-variant situation; Character recognition; Color; Euclidean distance; Humans; Image reconstruction; Image segmentation; Object recognition; Pattern recognition; Target recognition; Testing;
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.548984
