A modified fuzzy C-means image segmentation algorithm for use with uneven illumination patterns
Ma, Li and Staunton, R. C.. (2007) A modified fuzzy C-means image segmentation algorithm for use with uneven illumination patterns. Pattern Recognition, Vol.40 (No.11). pp. 3005-3011. ISSN 0031-3203Full text not available from this repository.
Official URL: http://dx.doi.org/10.1016/j.patcog.2007.02.005
A novel fuzzy C-mean (FCM) algorithm is proposed for use when active or structured light patterns are projected onto a scene. The underlying inhomogeneous illumination intensity due to the point source nature of the projection, surface orientation and curvature has been estimated and its effect on the object segmentation minimized. Firstly, we modified the recursive FCM algorithm to include biased illumination field estimation. New clustering center and fuzzy clustering functions resulted based on the intensity and average intensity of a pixel neighborhood based object function. Finally, a dilation operator was used on the initial segmented image for further refinement. Experimental results showed the proposed method was effective for segmenting images illuminated by patterns containing underlying biased intensity fields. A higher accuracy was obtained than for traditional FCM and thresholding techniques.
|Item Type:||Journal Article|
|Subjects:||Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software
T Technology > TK Electrical engineering. Electronics Nuclear engineering
|Divisions:||Faculty of Science > Engineering|
|Journal or Publication Title:||Pattern Recognition|
|Number of Pages:||7|
|Page Range:||pp. 3005-3011|
|Access rights to Published version:||Restricted or Subscription Access|
Actions (login required)