DocumentCode :
2403641
Title :
Fuzzy information system based on image segmentation by using shadow detection
Author :
Muthukumar, Shivkumar ; Krishnan, Nikhil ; Nachiyar, K.T. ; Pasupathi, P. ; Deepa, S.
Author_Institution :
Centre for Inf. Technol. & Eng., MS Univ., Tirunelveli, India
fYear :
2010
fDate :
28-29 Dec. 2010
Firstpage :
1
Lastpage :
6
Abstract :
Shadows arise when objects completely or somewhat occlude direct light from a source of illumination. This method is based on Tri color attenuation model (TAM). The proposed approach is based on shadow properties and exploits color information. Shadow is treated as a extraordinary kind of image degradation. Shadow area generally has lower brightness than non-shadow region. Based on TAM a shadow detection algorithm is presented here. For the segmentation intention a fuzzy c-means algorithm is adopted. Shadow edge detection is conducted based on the color dependability. They derive scene texture edges from an image and by subtracting those edges from the raw input; they obtain a shadow free color image This algorithm can be applied to real difficult outdoor scenes to detect shadows exclusive of any aforementioned knowledge. The experimental results are shown that this method achieve robustness and high segmentation accurateness.
Keywords :
edge detection; fuzzy set theory; image colour analysis; image segmentation; image texture; natural scenes; TAM; color information; fuzzy C-means algorithm; fuzzy information system; image degradation; image segmentation; image texture; occlusion; shadow edge detection; tri color attenuation model; Algorithm design and analysis; Clustering algorithms; Image color analysis; Image segmentation; Light sources; Partitioning algorithms; Pixel; Clustering; Fuzzy C-mean; GA; TAM; TS;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Computing Research (ICCIC), 2010 IEEE International Conference on
Conference_Location :
Coimbatore
Print_ISBN :
978-1-4244-5965-0
Electronic_ISBN :
978-1-4244-5967-4
Type :
conf
DOI :
10.1109/ICCIC.2010.5705906
Filename :
5705906
Link To Document :
بازگشت