Registration of Infrared Image and Digital Graphic Based Feature

Aiming to the characteristics of airborne infrared image and digital raster graphic, a hierarchical registration method based on feature is proposed. Firstly, the perspective distortion and rotation of the infrared image are corrected. Secondly, the log-polar technique based on edge features is implemented in order to achieve the preliminary registration result. Thirdly, matching points are selected by the relaxation iteration algorithm which using distance-ration invariant in detail. Finally, the precise registration result is calculated by the least square algorithm. The experimental results shows that the proposed algorithm is robust with high-precision, so it can meet the needs of aided-to-navigation.