Visual efficiency of image fusion methods
article
The objective of image fusion is to represent relevant information from multiple individual images in a single image. Some fusion methods may represent important visual information more distinctively than others, thereby conveying it more efficiently to the human observer. Here, we propose to rank order images fused by different methods according to the computational attention value of their regions of interest (ROI). To achieve this aim, we compute a multi-bitrate attention map for each of the fused images, following a rational model of computational attention. From this attention map, we then calculate the average attention score within areas of interest for each bitrate while using a prioritisation scheme. Here, the prioritisation protocol is used to simulate the basic cognitive process of visual information acquisition by human users. A huge computed mean attention value within the areas of interest at any reconstruction fidelity corresponds to a high-computational saliency of the areas of interest. The novelty of our approach to image fusion evaluation is the use of rate-attention curves, which are given by the normalised mean attention score within the areas of interest across bitrates while using a prioritisation protocol, and where the detection of ROI is achieved either interactively through human user intervention or by the use of automated detection. © 2012 Copyright Taylor and Francis Group, LLC.
Topics
TNO Identifier
446873
ISSN
19479832
Source
International Journal of Image and Data Fusion, 3(1), pp. 39-69.
Pages
39-69
Files
To receive the publication files, please send an e-mail request to TNO Repository.