An evaluation issue of image fusion method based on SVD (singular value decomposition) is put forward aiming at the inconsistencies among the evaluation conclusions drawn from analyses with different methods. In this way, the SVD is done for the numerical matrix composed of multiple inconsistency conclusions. Then, an approximate conclusion matrix is given by equilibrating the consistency-reliability indices. The method has the advantages of picking up common information from multi-evaluations, weakening the impact due to extreme evaluation conclusions. Experimental analysis shows that the proposed method performs well on validity, real-time, precision, and objectivity.%针对图像融合效果中存在的”多评价结论非—致性”问题,提出了一种基于奇异值分解的图像融合效果综合评价方法.首先对多个性能指标组成的序值矩阵进行奇异值分解,并运用一致可信度指标确定序值矩阵的近似矩阵,而近似矩阵是对原有矩阵的优化.该方法具有提取多评价结论共性信息,削弱极端评价结论影响,评价结果易于定量表示,更加精确、客观,区分度大,可靠性高等优点.实验结果表明,该评价方法具有较好的实时性和准确性,进一步丰富和完善图像融合理论框架具有启发意义和实用价值.
展开▼
