A Remote Sensing Image Target Recognition Method Based on SIFT Algorithm

Abstract. Aerial camera could obtain the ground target image information, and the target recognition could be carried through the image preprocessing and interpretation. In the process of image interpretation, under the precondition of massive images, if the interpretation member recognition was used only, the interpretation process will be complicated, time-consuming and the man-made errors will also be increased gradually. Based on this situation the SIFT method was used in the paper and the target in image received from aerial camera were got feature extraction, the feature points were obtained and it were contrast to the target feature points of standard target in the library which were set up in advance, so the accurate interpretation of aviation image target could achieved, so a kind of effective interpretation identification method for correct aviation image target recognition was proposed