Parallel image matching for aircraft navigation

Image matching is an important technique for aircraft navigation. This paper presents a novel parallel image matching algorithm based on local entropy vector (LEV). Spatial information is introduced into the definition of image entropy and the proposed algorithm confirms the robustness of LEV to irradiation distortion, geometric distortion, local noise interference and partial occlusion. An equivalent transform of the image entropy definition is also employed to make it suitable for parallel processing. The matching results of the proposed algorithm are validated by experiments on a variety of images in natural environment.