Title :
Fusion Image Quality Assessment Based on Quaternion
Author :
Rui Zhang ; Ting Jiang ; Yao-yao Yu ; Hui Gong ; Guang-jun Dong
Author_Institution :
Inst. of Surveying & Mapping, Inf. Eng. Univ., Zhengzhou, China
Abstract :
A new fusion image quality assessment method according with human vision system and quaternion was introduced in this paper. After encoded RGB channels into three imaginary parts of quaternions, the matrix of pure quaternion, which with null real part for every element, can describe color image transactions as some kind of transformations, so that the entire color information of image was preserved. Treating multispectral image to be fused as reference image, fusion image as image to be assessed, both of two images were executed by singular value decomposition. Then fusion image quality assessment having referenced information was realized via distance and angle between the two singular vectors. The tests results show that the method mentioned in this paper have the consistency of objectivity and subjectivity, and can not only verity human visual system, and the definition and spectrum retain-ability capability of fusion image can be evaluate also.
Keywords :
image colour analysis; image fusion; singular value decomposition; RGB channel; color image transaction; color information; fusion image quality assessment; human vision system; human visual system; multispectral image; pure quaternion; reference image; singular value decomposition; Color; Equations; Humans; Image color analysis; Image fusion; Image quality; Quaternions;
Conference_Titel :
Image and Data Fusion (ISIDF), 2011 International Symposium on
Conference_Location :
Tengchong, Yunnan
Print_ISBN :
978-1-4577-0967-8
DOI :
10.1109/ISIDF.2011.6024302