A Novel and Better Performance Evaluation Algorithm for Image Fusion Based on Hierarchical Grey Relational Analysis
Guiqing He,Hua Qi,Yi Wang,Li Guan,Chongyang Hao
DOI: https://doi.org/10.3969/j.issn.1000-2758.2007.01.027
2007-01-01
Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University
Abstract:Aim. To our knowledge, there is as yet no paper in the open literature dealing with performance evaluation of image fusion based on hierarchical grey relational analysis. We now present such performance evaluation, which is not only novel but also we have come to regard as being highly promising to make such performance evaluation, better in several ways. In the full paper, we explain our algorithm and its application in detail; in this abstract, we just add some pertinent remarks to listing the two topics of explanation: (1) the common methods of evaluating image fusion performance and (2) performance evaluation algorithm for image fusion based on hierachical grey relational analysis; the three subtopics of topic 2 are grey relational analysis (subtopic 2.1), the performance evaluation algorithm for image fusion based on hierarchical grey relational analysis (subtopic 2.2) and experimental results and analysis (subtopic 2.3); in subtopic 2.2, we first give a six-step procedure for performance evaluation of a single level of hierarchical grey relational analysis and then, based on the single level analysis, we give a four-step procedure for performance evaluation of hierarchical grey relational analysis; in subtopic 2.3, the careful analysis of the results, summarized in Tables 2 and 3 in the full paper respectively for two experiments, appears to show that our algorithm gives performance evaluation more precise and objective than those given by other algorithms; the above-mentioned careful analysis also shows that our algorithm gives not only overall performance but also required special performance evaluation; still in subtopic 2.3, we point out that our algorithm can complete performance evaluation in milliseconds and thus it is capable of selecting the method of image fusion suitable for the fusion of the particular images at the moment.