Abstract:

Similarity measure is a very important problem in image registration. Three typical similarity measures for remote sensing image registration are selected including mutual information, correlation coefficient and sum of squared differences. Then, the performances of these measures are compared and ana lyzed by the computing time,the sharpness,the sensitivity of noise and the effect on multimodal image registration. The results of experiments show that  these similarity measures have different effects and performances in different application circumstances.

Key words: remote sensing image, image registration, mutual information, correlation coefficient, sum of squareddifferences