CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
Object recognition in machine vision system of mobile robot: the using of library “FLANN” and algorithm “SURF”
 |
Read | Nguen Tuan Zung , Shcherbatov Ivan A. Object recognition in machine vision system of mobile robot: the using of library “FLANN” and algorithm “SURF” // Caspian journal : management and high technologies. — 2014. — №4. — pp. 65-76. |
Nguen Tuan Zung - post-graduate student, Astrakhan State Technical University, 16 Tatishchev St., Astrakhan, 414056, Russian Federation, dunghvkt@yahoo.com
Shcherbatov Ivan A. - Ph.D. (Engineering), Associate Professor, Astrakhan State Technical University, 16 Tatishchev St., Astrakhan, 414056, Russian Federation, sherbatov2004@mail.ru
Object detection (obstacles detection) in the near field of mobile robot is very urgent and difficult task. In this paper, we propose a method, which uses nearest neighbor search В«FLANN - fast library for approximate nearest neighborsВ», and algorithm В«SURFВ» to find matching images in the collection of images. Library В«FLANNВ» contains a collection of algorithms: extracting descriptors of the image, computing descriptors for multiple images, create index FLANN, performing searches KNN (k-nearest neighbor method), and getting results. Efficiency of finding objects will increase with the use of the algorithm SURF. Using library FLANN conjunction with SURF algorithm meets the requirements for detection of objects in real time. The experimental results show the effectiveness of the proposed approach when used in the vision system of mobile robot platform.
Key words: библиотека FLANN, алгоритм SURF, SIFT, мобильный робот, распознавание изображений, K-мерное дерево, поиск ближайшего соседа, the FLANN library, SURF algorithm, SIFT, mobile robot, image recognition, K-dimensional tree, the nearest neighbor search