Automatic data classification
Within this topic we want to learn, to consolidate and to develop in depth the techniques of image and cloud processing to classify elements both semantically and non-semantically. The aim of this study is twofold: 1) to use this information as input information in positioning algorithms; 2) to be able to create useful databases for technitians and public administrations.
We understand by non-semantic classification that one based on the analysis of the elements of the environment without taking into account its entity, that is to say, we are interested in its geometrical properties but not on its identity.
Figure: Automatic road signal extraction
P. Kumar, P. Lewis, Automated extraction of road median from airborne laser scanning data , in Proceedings of 38th Asian Conference on Remote Sensing (ACRS 2017), 23-27 October 2017, New Delhi (India).
P. Kumar, P. Lewis, T. McCarthy, The potential of active contour models in extracting road edges from mobile laser scanning data , Infrastructures, Vol. 2, No. 3, pp. 16, July 2017.
P. Kumar, P. Lewis, C. P. McElhinney, P. Boguslawski, T. McCarthy, Snake energy analysis and results validation for a mobile laser scanning data based automated road edge extraction algorithm , IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 10, No. 2, pp. 763-773, February 2017, http://ieeexplore.ieee.org/document/7473830/.
E. Angelats, M. E. Parés, I. Colomina, The potential of non-semantic features for UAV remote sensing data fusion , in Proceedings of 2nd Virtual Geoscience Conference, 21–23 September 2016, Bergen (Norway).