基于逼近理想解排序的图像融合效果评价方法

将图像融合评价看作是一个多属性决策问题, 提出了一种基于逼近理想解排序的图像融合效果综合评价方法。通过引入理想融合图像、负理想融合图像的概念, 结合评价者的先验知识和选取的定量评价指标, 以距离测度作为图像融合结果的优劣评判标准，实现了对各个融合图像效果的综合评价, 兼顾了实际评价工作中的通用性和特殊性要求。实验结果表明, 该评价方法具有较好的有效性、实时性和简洁性。

Performance evaluation approach to image fusion based on ordering approximate to ideal solution

The evaluation issue of fused images is regarded as a multi attribute decision make problem, thus a novel synthesis approach using the technique for ordering approximate to ideal solution is proposed. By introducing the concept of ideal fused image, negative ideal fused image, combining the estimator’s prior knowledge with quantitative evaluation index, and taking distance measurement as the evaluation criterion for different fused images, the overall performance evaluation of the fused images is implemented, which gives not only universal performance but also required special performance evaluation. Experimental analysis shows that the proposed method performs well on validity, real time, precision, objectivity and versatility.