DocumentCode :
352447
Title :
A low cost hybrid diffusion technique for object segmentation
Author :
Izquierdo, Ebroul ; Ghanbari, Mohammed
Author_Institution :
Dept. of Electron. Syst. Eng., Essex Univ., Colchester, UK
Volume :
6
fYear :
2000
fDate :
2000
Firstpage :
2251
Abstract :
In this paper a low-complexity nonlinear filtering technique to smooth textures preserving object contours is presented. The approach is based on a hybrid combination of both isotropic and anisotropic recursive filtering. Using only intensity information the segment borders obtained by applying nonlinear filtering do not necessarily coincide with physical object contours, especially in the case of textured objects. To segment images into regions with physical meaning additional information extracted from disparity or motion is used to weight the filter coefficients. The presented technique has been successfully tested in the context of object segmentation of natural scenes and object-based disparity estimation for stereoscopic applications
Keywords :
feature extraction; image segmentation; image texture; nonlinear filters; recursive filters; smoothing methods; stereo image processing; anisotropic recursive filtering; filter coefficients weighting; image segmentation; intensity information; isotropic recursive filtering; low cost hybrid diffusion technique; low-complexity nonlinear filtering technique; natural scenes; object contours preservation; object segmentation; object-based disparity estimation; segment borders; stereoscopic applications; textured objects; textures smoothing; Anisotropic magnetoresistance; Costs; Data mining; Image processing; Image segmentation; Information filtering; Information filters; Layout; Object segmentation; Shape;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on
Conference_Location :
Istanbul
ISSN :
1520-6149
Print_ISBN :
0-7803-6293-4
Type :
conf
DOI :
10.1109/ICASSP.2000.859287
Filename :
859287
Link To Document :
بازگشت