Title:	Navigation parameters correction technique using multiple view geometry methods
Authors:	Sablina, Victoria A. Novikov, Anatoly I. Nikiforov, Michael B. Loginov, Alexander A.
Citation:	WSCG 2014: communication papers proceedings: 22nd International Conference in Central Europeon Computer Graphics, Visualization and Computer Visionin co-operation with EUROGRAPHICS Association, p. 103-110.
Issue Date:	2014
Publisher:	Václav Skala - UNION Agency
Document type:	konferenční příspěvek conferenceObject
URI:	wscg.zcu.cz/WSCG2014/!!_2014-WSCG-Communication.pdf http://hdl.handle.net/11025/26384
ISBN:	978-80-86943-71-8
Keywords:	multispektrální systém počítačového vidění;geometrie více zobrazení;analýza obrazů;korekce navigačních parametrů;detekce objektů;homografie;letadlo
Keywords in different language:	multispectral computer vision system;multiple view geometry;contour analysis;navigation parameters correction;object detection;homography;aircraft
Abstract in different language:	The problem of determining precisely the navigation parameters of an aircraft arise when constructing a multispectral computer vision system. In particular, these parameters are used to solve the image superimposition task. It is necessary to superimpose the aerial photograph registered from the on-board sensors and the image generated from the digital terrain map. Such superimposition will be successful only if this image is synthetized for the viewpoint based on the correct navigation parameters. However, the navigation parameters obtained from satellite and inertial navigation systems have inadmissible errors for this task. Thus it is required to reduce the influence of the errors by using the information from the analyzed images themselves. In this paper we suggest the technique to correct the navigation parameters in such a way. The technique is based on the application of the contour analysis and multiple view geometry methods. These methods make it possible to implement our technique in real time due to the low computational complexity of the contour analysis and multiple view geometry algorithms. The results represented in the paper indicate that the further development and application of the suggested technique for multispectral computer vision systems are promising.
Rights:	@ Václav Skala - UNION Agency
Appears in Collections:	WSCG 2014: Communication Papers Proceedings

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.