Paper
15 October 2013 Image structural analysis in the tasks of automatic navigation of unmanned vehicles and inspection of Earth surface
Vadim Lutsiv, Igor Malyshev
Author Affiliations +
Abstract
The automatic analysis of images of terrain is urgent for several decades. On the one hand, such analysis is a base of automatic navigation of unmanned vehicles. On the other hand, the amount of information transferred to the Earth by modern video-sensors increases, thus a preliminary classification of such data by onboard computer becomes urgent. We developed an object-independent approach to structural analysis of images. While creating the methods of image structural description, we did our best to abstract away from the partial peculiarities of scenes. Only the most general limitations were taken into account, that were derived from the laws of organization of observable environment and from the properties of image formation systems. The practical application of this theoretic approach enables reliable matching the aerospace photographs acquired from differing aspect angles, in different day-time and seasons by sensors of differing types. The aerospace photographs can be matched even with the geographic maps. The developed approach enabled solving the tasks of automatic navigation of unmanned vehicles. The signs of changes and catastrophes can be detected by means of matching and comparison of aerospace photographs acquired at different time. We present the theoretical proofs of chosen strategy of structural description and matching of images. Several examples of matching of acquired images with template pictures and maps of terrain are shown within the frameworks of navigation of unmanned vehicles or detection of signs of disasters.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Vadim Lutsiv and Igor Malyshev "Image structural analysis in the tasks of automatic navigation of unmanned vehicles and inspection of Earth surface", Proc. SPIE 8897, Electro-Optical Remote Sensing, Photonic Technologies, and Applications VII; and Military Applications in Hyperspectral Imaging and High Spatial Resolution Sensing, 88970F (15 October 2013); https://doi.org/10.1117/12.2028840
Lens.org Logo
CITATIONS
Cited by 6 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image analysis

Photography

Transform theory

Unmanned vehicles

Aerospace engineering

Robots

Sensors

Back to Top