SMA: A Suitable-Matching Area Selection Dataset

Scene matching is a commonly used guidance technology in the final guidance stage of missiles. Suitable-matching area selection is the core technology of reference map preparation and one of the critical factors affecting the performance of the scene-matching system. Since there is no public dataset for scene-matching suitable area selection, it is difficult to compare related selection algorithms uniformly. In this paper, we analyze the scene categories that may be encountered in the scene-matching process and create a dataset SMA, which contains six scene types and various scene contents. This dataset can be used for matching performance analysis and evaluation of scene areas. We analyze the information content, stability, uniqueness, and significance of the images and conduct matching experiments by simulating real-time images to prove the matching performance of the calibration area.