Compressive sensing with local geometric features
Proceedings of the twenty-seventh annual symposium on Computational geometry
| Hi-index | 0.00 |
We present here a new algorithm which enables the identification of stars in astronomical photographs using readily available star catalogs for this purpose. The algorithm was implemented in a stand-alone application called `Skyprint` capable of performing a matching process. The computational aspect of the problem can be designated to the wide class of image recognition methods and analysis of multidimensional data. The astronomical aspect concentrates on astrometry --- the method of determining the coordinates of stars in the celestial sphere. The problem of identifying star patterns occurs most often in such areas as cosmic probe navigation, adjusting and merging numerous photographs of the sky together, or in recovering missing information in relation to a fragment of the sky represented in the photograph.