Objective image fusion evaluation method for target recognition based on target quality factor

A novel objective evaluation method for image fusion based on target quality factor is proposed from the point of view of target recognition. Three-component indicators named preservation degree of special features of the target, target edge quality and interference edge suppression ratio are defined. These component indicators can evaluate the quality of the target from three different aspects. Preservation degree of special features of the target quantitatively describes the preservation degree of special features in the fusion image, which are very important for target recognition in the special source images. Target edge quality is used to evaluate the integrity and quality of the boundary information contained in the fusion image. Interference edge suppression ratio quantitatively describes the boundary information whether the fusion image will produce confusion. Target quality factor is obtained by weighted averaging of these component indicators. The experimental results show that the target quality factor can evaluate the image fusion for target recognition quantitatively and reasonably, and the evaluation results are in accordance with the visual effect of the fusion images.

Automated summary

A novel objective evaluation method for image fusion based on target quality factor is proposed, which can quantitatively and reasonably evaluate the image fusion for target recognition. The evaluation results are in accordance with the visual effect of the fusion images.