13 October 2015 Automatic structural matching of 3D image data
Author Affiliations +
Abstract
A new image matching technique is described. It is implemented as an object-independent hierarchical structural juxtaposition algorithm based on an alphabet of simple object-independent contour structural elements. The structural matching applied implements an optimized method of walking through a truncated tree of all possible juxtapositions of two sets of structural elements. The algorithm was initially developed for dealing with 2D images such as the aerospace photographs, and it turned out to be sufficiently robust and reliable for matching successfully the pictures of natural landscapes taken in differing seasons from differing aspect angles by differing sensors (the visible optical, IR, and SAR pictures, as well as the depth maps and geographical vector-type maps). At present (in the reported version), the algorithm is enhanced based on additional use of information on third spatial coordinates of observed points of object surfaces. Thus, it is now capable of matching the images of 3D scenes in the tasks of automatic navigation of extremely low flying unmanned vehicles or autonomous terrestrial robots. The basic principles of 3D structural description and matching of images are described, and the examples of image matching are presented.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Svjatoslav Ponomarev, Svjatoslav Ponomarev, Vadim Lutsiv, Vadim Lutsiv, Igor Malyshev, Igor Malyshev, } "Automatic structural matching of 3D image data", Proc. SPIE 9649, Electro-Optical Remote Sensing, Photonic Technologies, and Applications IX, 96490M (13 October 2015); doi: 10.1117/12.2194312; https://doi.org/10.1117/12.2194312
PROCEEDINGS
13 PAGES


SHARE
Back to Top