Navigation Landmark Recognition and Matching Algorithm Based on the Improved SURF

The satellite measurement landmark information is a new autonomous navigation information source which can provide a new method to improve the accuracy of satellite autonomous orbit determination. The landmark image recognition and matching performance is one of the key factors affecting the measurement accuracy. Aiming at the application limitations of the traditional speeded up robust features (SURF) algorithm in image matching, an improved SURF algorithm is proposed combined with image preprocessing technique, Harris feature point detection method and mismatching point elimination technique to recognize and match the landmark. Harris feature point detection method can improve the feature points extraction stability and the robustness of affine transformation. The low pass filter is used to preprocess the image to reduce the influence of noise on the recognition. The relative distance method is used to eliminate the mismatching points. According to the satellite camera parameters, Bohai Bay, Zhaling Lake and Penghu Island are chosen as the images to be matched. Then the matching points and the matching accuracy are chosen as the indicators to verify the performance of the proposed algorithm. The simulation results indicate that the matching accuracy of the traditional SURF algorithm is over 70% and the matching accuracy of the improved SURF algorithm is over 90%.