Satellite Image Registration Using Nature Inspired Techniques

Automatic satellite image registration for multi-sensor images is becoming increasingly important to aid in flood damage assessment. We consider two images, the one before flood (optical image) and the other during flood (SAR image) in the registration process. The objective is to maximize the similarity metric (of these two images) using information theoretic measures such as Mutual Information (MI). The maximum MI would imply that the images are registered better. The function of these metric for transformation parameters are generally non-convex and irregular and, therefore, makes it difficult to use the standard optimization methods for the global solution. In this paper, we present the nature inspired techniques - Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Artificial Bee Colony (ABC) to search the maximum MI. From the results obtained, we compare the performance evaluation and conclude that the nature inspired techniques are accurate and reliable in solving the automatic satellite image registration.