Using an Imperialistic Competitive Algorithm in Global Polynomials Optimization (Case Study: 2D Geometric Correction of IKONOS and SPOT Imagery)

The number of high resolution space imageries in photogrammetry and remote sensing society is growing fast. Although these images provide rich data, the lack of sensor calibration information and ephemeris data does not allow the users to apply precise physical models to establish the functional relationship between image space and object space. As an alternative solution, some generalized models such as global polynomials have been developed and used. This paper presents a hybrid method based on using imperialistic competitive algorithm (ICA) to find the best terms of global polynomials. The method was carried out for geometric correction of two different datasets, an IKONOS Geo-image and a SPOT image, with different number of ground control points (GCPs) and independent check points (ICPs). Results showed the success of achieving sub-pixel accuracies (0.2) for IKONOS and 2.5 pixels for SPOT image. The method was able to successfully handle over-optimization as it produces lower RMSEs compared to conventional approach. Also, the proposed method required much less time in comparison to other optimization algorithms like genetic algorithm (GA) and particle swarm optimization (PSO).