Performance Measures for Fusion of Multiple Input Images

This paper discusses three performance measures that can handle fusion techniques, using which; multiple input images can be fused. The evaluation methods are based on the degree of correlation and the deviation of information between the source images and the fused image. The performance measures are: Nonlinear Correlation Coefficient (NCC), Nonlinear Correlation Information Entropy (NCIE) and Information Deviation Analysis (IDA). Most of the quality assessment metrics used for evaluating fusion algorithms that has been presented in literature is applicable only when there are two input images. However, there are cases, where we need to fuse multiple (more than two) input images. In this paper, four spatial domain fusion algorithms applied for multiple input images are compared and analyzed, based on NCC, NCIE and IDA.