Title :
Automatic extraction of region of interest from images with low depth of field
Author :
Kim, MinChul ; Jeong, Hong
Author_Institution :
Pohang Univ. of Sci. & Technol., Pohang, South Korea
Abstract :
In this paper, we propose a new segmentation algorithm for separate focused foreground region from defocused background region with low depth of field image. This algorithm first computes the blur parameter each of pixel. Then, these parameter values are clustered to blurry regions and sharp regions using belief propagation algorithm. Experimental results show that this approach also can separate the focused image from defocused image.
Keywords :
belief maintenance; feature extraction; image segmentation; pattern clustering; belief propagation algorithm; blur parameter; blurry region clustering; defocused background region; depth-of-field image; image extraction; region-of-interest image; segmentation algorithm; separate focused foreground region; Belief propagation; Cameras; Clustering algorithms; Data mining; Focusing; Geometrical optics; Image segmentation; Information technology; Lenses; Solid modeling; defocus image; image segmentation; low depth of field;
Conference_Titel :
Networking and Information Technology (ICNIT), 2010 International Conference on
Conference_Location :
Manila
Print_ISBN :
978-1-4244-7579-7
Electronic_ISBN :
978-1-4244-7578-0
DOI :
10.1109/ICNIT.2010.5508518
