Scholarly record
IDENTIFICATION OF SPATIAL OBJECTS WITH THE SAME TOPOLOGICAL STRUCTURE IN IMAGES FROM DIFFERENT TIMES OF THE YEAR
Abstract
Currently every day the need for obtaining information from satellites and quadcopters is increasing. Various services and ordinary users have a need for automated processing of received information from satellites. However, automated processing raises a number of problems: different illumination of objects, their difference in different times for example at different times of the year, soil changes, river spills and forests before and after fires etc. However, topology allows you to preserve the shape of feature data. The article sets the task of developing an algorithm for searching for spatial objects with the same topological structure for different seasons. The spatial objects identification algorithm based on topological features, which using barcodes formed using simplicial complexes and persistent homology. The search takes place on various satellite images. The result of the algorithm is the detection of objects in images for different time intervals based on the obtained three-dimensional barcodes. The analysis of various approaches to the search for spatial objects in satellite images is carried out the advantages and disadvantages are identified. In the work images of rivers and lakes were studied for different years and in the spring, winter, autumn and summer seasons. Plots of topological characteristics of these objects are constructed. Similarities and differences of topological characteristics are displayed. The developed algorithm allows you to find spatial objects with an accuracy of 90%. In the future the algorithm can be improved by introducing additional topological features and the accuracy of determination can be about 95%. This algorithm can be applied not only to rivers and to lakes but when searching for changes in soils, forests or when gas explosions appear in the Arctic.
Publication Impact Profile
- Captures
- Mendeley - Readers: 1
Publication details
References5
Andrianov D. E., Eremeev S. V., Kovalev. Y. A., Kuptsov K. V. Algorithm for encoding nD spatial objects into GIS // CEUR Workshop Proceedings. Image Processing and Earth Remote Sensing. Information Technology and Nanotechnology. – 2018. – Vol. 2210. – Pp. 149-155.
Eremeev S. V., Kovalev. Y. A., Kuptsov K. V. A research of classification algorithm of spatial information on the basis of methods of persistent homology and random forest // CEUR Workshop Proceedings. Data Science. Information Technology and Nanotechnology 2018. – Vol. 2212. – Pp. 165- 171.
Zhou G., Wang Y. Occlusion detection for urban aerial true orthoimage generation // IEEE International Geoscience and Remote Sensing Symposium (IGARSS). – 2016. – P. 3009-3012.
Gioi R. G. et al. LSD: A fast line segment detector with a false detection control // IEEE transactions on pattern analysis and machine intelligence. – 2010. – Vol. 32. – № 4. – P. 722-732. . John Wiley. 3. Haining, R. Spatial Data Analysis. 2003. P. 430.
Phuoc Vinh Tran. Overview of Three and Four-Dimensional GIS Data Models. // Lecture Notes in Electrical Engineering. 2013. Pp. 1013-1020.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.