Constructing and visualizing three-dimensional sea bottom models to test AUV machine vision systems
- Authors: Kamaev A.N.1, Sukhenko V.A.2, Karmanov D.A.1
-
Affiliations:
- Computing Center of the Far Eastern Branch of the Russian Academy of Sciences
- Biometric Technologies
- Issue: Vol 43, No 3 (2017)
- Pages: 184-195
- Section: Article
- URL: https://journals.rcsi.science/0361-7688/article/view/176507
- DOI: https://doi.org/10.1134/S0361768817030070
- ID: 176507
Cite item
Abstract
This paper describes an algorithm for constructing a procedural sea bottom model, which can be used for testing and debugging machine vision systems of autonomous underwater vehicles (AUVs). The algorithm consists of three main stages: generating a low-frequency heightmap (used by the designer to define the basic form of a water area), constructing a three-dimensional model (based on the heightmap and fractal noise), and visualizing the three-dimensional model (refined by means of hardware or manual tessellation). The sea bottom model has the following features: it is detailed accurate to a screen pixel, each of its sections is absolutely unique, and its size is adequate for any tests.
About the authors
A. N. Kamaev
Computing Center of the Far Eastern Branch of the Russian Academy of Sciences
Author for correspondence.
Email: kamaev_an@mail.ru
Russian Federation, ul. Kim Yu Chena 65, Khabarovsk, 680000
V. A. Sukhenko
Biometric Technologies
Email: kamaev_an@mail.ru
Russian Federation, ul. Beloostrovskaya 17/2, St. Petersburg, 197342
D. A. Karmanov
Computing Center of the Far Eastern Branch of the Russian Academy of Sciences
Email: kamaev_an@mail.ru
Russian Federation, ul. Kim Yu Chena 65, Khabarovsk, 680000