Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2013, Vol. 10, No. 4, pp. 98-106
A method of computational optimization for matching vector and raster remote sensing data
V.G. Bondur
1, A.B. Murynin
2, I.A. Matveev
2, A.N. Trekin
3, I.A. Yudin
11 ”AEROCOSMOS” Institute for Scientific Research of Aerospace Monitoring, Moscow, Russia
2 ”AEROCOSMOS” Institute for Scientific Research of Aerospace Monitoring; Dorodnicyn Computing Centre of Russian Academy of Sciences, Moscow, Russia; Moscow, Russia
3 ”AEROCOSMOS” Institute for Scientific Research of Aerospace Monitoring; Moscow Institute of Physics and Technology (State University), Moscow, Russia; Dolgoprudny, Moscow rgn., Russia
Various tasks of remote sensing require combined processing of raster and vector data of the territory. Thus the problem of matching geo-referenced raster against polygons in geographical coordinates arises. Simple method of spatial indexation of the vector polygons is proposed, which increases the speed of vector data search and facilitates the matching. The algorithm has been implemented and tested with real remote sensing data and its performance was compared to analogue. Algorithmic complexity and the way of setting optimal parameters of indexation are analyzed theoretically and confirmed by experiments.
Keywords: индексация, обработка изображений, дистанционное зондирование Земли, spatial indexing, image processing, remote sensing
Full text