Translator Disclaimer
4 May 2010 Blind quality assessment of multi-focus image fusion algorithms
Author Affiliations +
At present time, image fusion is widely recognized as an important aspect of information processing. It consists of combining information originated from several sources in order to improve the decision making process. In particular, multi-focus image fusion combines images that depict the same scene but they are not in-focus everywhere. The task seeks to reconstruct an image as sharp as possible by preserving in-focus areas while discarding blurred areas. The quality of fused images is of fundamental importance. Many objective quality metrics for image fusion have been proposed. However, the evaluation of fused images is still a difficult task, especially because there is no reference image to compare with. Blind image quality assessment refers to the problem of evaluating the visual quality of an image without any reference. In this paper, we describe a blind image fusion quality assessment procedure based on the use of mutual information (MI). This procedure is concise and explicit and will be useful in scenarios where the absence of a reference image can hamper the assessment of the results. Furthermore, several image fusion algorithms have been rated and they have shown that our metric is compliant with subjective evaluations. Consequently, it can be used to compare different image fusion methods or to optimize the parameter settings for a given fusion algorithm.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Rodrigo Nava, Boris Escalante-Ramírez, and Gabriel Cristóbal "Blind quality assessment of multi-focus image fusion algorithms", Proc. SPIE 7723, Optics, Photonics, and Digital Technologies for Multimedia Applications, 77230F (4 May 2010);

Back to Top