30 October 2009 Image fusion based on wavelet transform and visual attention mechanism
Author Affiliations +
Proceedings Volume 7494, MIPPR 2009: Multispectral Image Acquisition and Processing; 74941X (2009) https://doi.org/10.1117/12.833235
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
Image fusion is a process of combing multiple images of the same scene into a single image with the aim to preserve the full content information and retain the important features from each of the combined images. In this paper, a novel image fusion method based on Wavelet Transform (WT) and Visual Attention Mechanism (VAM) is proposed. Firstly, the source images are decomposed by WT to get the sub-images. Secondly, by using the VAM, the salience maps of the source images are formed, which can indicate the salient regions to the human visual system, i.e., the higher the saliency value is, the more important the location represent. The saliency maps are then used together with the match measure between the coefficients to guide the combination of the coefficients as follows: if the match measure at a given location is low, the coefficient from the source image with higher saliency is selected to be the fused coefficient. Contrarily, if the match measure at a given location is high, then the fused coefficients are calculated as the weighted sum of the coefficients extracted from both source images. The weights here are determined by the corresponding saliency maps of the source images. Finally, the fusion image is obtained by using the inverse WT transform. Experimental results applying the proposed algorithm show that the fused image keeps more visual meaningful information than other methods.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xianyi Ren, Wei Liu, Peng Duan, Sanfeng Chen, "Image fusion based on wavelet transform and visual attention mechanism", Proc. SPIE 7494, MIPPR 2009: Multispectral Image Acquisition and Processing, 74941X (30 October 2009); doi: 10.1117/12.833235; https://doi.org/10.1117/12.833235
PROCEEDINGS
6 PAGES


SHARE
Back to Top