GLCM-based metric for image fusion assessment

This paper introduces a novel metric for image fusion evaluation that is based on texture. Among the applications of image fusion are surveillance and remote sensing, where the combination of relevant information from multiple image sources are preserved in a fused output. From these, the conservation of background textural details is especially important as they help to define the image structure. The concept is brought forth in our work, which aims to evaluate the performance of image fusion algorithms based on their ability to retain textural details from the fusion process. We utilise the GLCM model to extract second-order statistical features for the derivation of an image textural measure. This is used to replace the edge-based calculations in the Petrovic metric. Performance evaluation on established fusion methods verify that the proposed metric is accurate, especially for multimodal scenarios.