作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 图形图像处理 • 上一篇    下一篇

图像融合算法的综合性能评价指标

陈立伟1,2,蒋 勇2   

  1. (1. 西南交通大学信息科学与技术学院,成都610031;2. 西南科技大学计算机科学与技术学院,四川绵阳621010)
  • 收稿日期:2014-01-15 出版日期:2015-02-15 发布日期:2015-02-13
  • 作者简介:陈立伟(1974 - ),男,副教授、博士研究生,主研方向:图形图像处理,光学模式匹配,智能信息化处理;蒋 勇,副教授、博士。
  • 基金资助:
    国家自然科学基金资助项目(10676029, 10776028);人工智能四川省重点实验室开放基金资助项目(2014RYY03)。

Comprehensive Performance Evaluation Index of Image Fusion Algorithm

CHEN Liwei 1,2,JIANG Yong 2   

  1. (1. School of Information Science and Technology,Southwest Jiaotong University,Chengdu 610031,China; 2. School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang 621010,China)
  • Received:2014-01-15 Online:2015-02-15 Published:2015-02-13

摘要: 图像融合算法性能评价是图像融合工程的重要组成部分,现有的融合评价指标从不同方面评价融合图像 质量,这些指标在评价图像融合算法性能时存在片面性,难以对融合算法的综合性能作出评价。为此,运用多指标 决策技术,提出一种加权总分指标,将多个指标评价值综合为单一值,从而对图像融合算法进行综合性能评价。将 加权总分指标的评价结果同逼近理想解排序指标以及秩和比指标的结果相比较,实验结果表明,该指标的评价结 果和主观评价结果一致,其综合评价能力与其他2 种指标相近,提高了融合算法综合性能评价结果的可靠性和准确性。

关键词: 图像融合, 加权总分指标, 融合性能, 多指标决策, 逼近理想解排序法, 秩和比

Abstract: Evaluating the overall performance of the image fusion algorithm is an important part of the image fusion project. The existing evaluation indexes of the fused image often evaluate the quality of the fused image from different aspects and thus have some one-sidednesses when they are used to evaluate the performance of the the fusion algorithm; it is difficult to evaluate the overall performance of the fusion algorithm. This paper,by employing the multi-index decision technique,proposes a Weighted Total Score(WTS) index. This index can synthesize multiple evaluation values into a single value and measure the overall performance of image fusion algorithm. It also compares the evaluation results of WTS with those of Technique for Order Preference by Similarity to Ideal Solution(TOPSIS) and Rank-sum Ratio (RSR). Experimental results show that WTS index can achieve the same results as those of the subjective evaluation and has the overall evaluation ability close to the other two indexes. The proposed index can improve the reliability and the accuracy of the evaluation of the overall performance of the different fusion algorithms.

Key words: image fusion, Weighted Total Score(WTS) index, fusion performance, multi-index decision, Technique for Order Preference by Similarity to Ideal Solution(TOPSIS), Rank-sum Ratio(RSR)

中图分类号: