Title :
A method for noise suppression in image morphing using co-occurrence frequency image
Author :
Nagasaka, Yosuke ; Fujiwara, Takayuki ; Funahashi, Takuma ; Koshimizu, Hiroyasu
Author_Institution :
Sch. of Inf. Sci. & Technol., Chukyo Univ., Toyota, Japan
Abstract :
This paper proposes a method for suppressing the noise in image morphing due to the incorrect correspondence between the texture edge borders. The proposed method was applied to the situation where there are some ill correspondences between input images. In the proposed method, the incorrectly corresponded region is detected by using the co-occurrence frequency image (CFI). A smart method for suppressing morphing noise was applied to the detected region. Experimental considerations and the coming subjects are presented to show the effectiveness of the proposed method.
Keywords :
image denoising; image morphing; image texture; cooccurrence frequency image; image morphing; noise suppression; texture edge borders; Feature extraction; Image color analysis; Image edge detection; Interpolation; Noise; Shape; Writing; Co-occurrence frequency image; Exterior division; Morphing; Noise suppression; Shape; Texture;
Conference_Titel :
Mechatronics and Automation (ICMA), 2011 International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-8113-2
DOI :
10.1109/ICMA.2011.5985734