Research On Automated Mosaic Method Of Aerial Image

Aerial Image has many perspective applications,such as early warning disaster control,digital map generation,military reconnaissance,target tracking and UAV-aided navigation etc. Automated Mosaic technology of Aerial Image play a key role.The characteristics of Aerial Image are different view points,large view scene.The difficulty of Automated Stitching of Aerial Image are detection and matching of the feature points and global optimization adjustment.Based on the analyses of the advantage and disadvantage of the method of area- based and of feature-based, we choose the method of SIFT to detection and matching. according to the characteristics of Aerial Image,we simplify the work flow of SIFT method and introduce the method of block-based for precision of the feature point. Due to the disperse and regular of view points and the view angle of Aerial Image,we introduce movement model which be apply for any change and use Nor-linear LMS method to estimating parameters. The cumulated-error is a inescapability problem for the mosaic of many images.based on the analyses the generation and transfer of the cumulated-error, the strategy of bundle adjustment and unitary optimization is a good solution.We introduce the method integration registration of with the restriction. The results of experiment prove that such algorithm is efficient and minish the less accumulating errors.