Journal of Engineering and Applied Sciences
ISSN: Online 1818-7803ISSN: Print 1816-949x
109
Views
1
Downloads
Abstract
Traffic density data plays important role in traffic management, road planning as well as urban land use planning. Several efforts have been used to gather this data, mainly by detecting and counting vehicles in roads by processing images from CCTV placed in certain positions in roads. The main disadvantage of this approach is that it is only possible to detect and count vehicles effectively in a relatively limited area of the roads due to limited height and camera resolution. By using satellite images or images taken from drones, the coverage area of the roads can be increased significantly, however problems of false detection due to objects looking similar to vehicles also increases. This reseach uses Template Matching methos by using correlation equation, haar cascade classification, keypoint detection using maximally stable extremal region and Oriented FAST and Rotated BRIEF (ORB) feature extraction method. The highest recall and precision value using MSER and ORB are 100 and 75%, respectively.
How to cite this article:
Joko Lianto Buliali, Chastine Fatichah, Darlis Herumurti, Diagnosa Fenomena, Hera Widyastuti and Mark Wallace. Vehicle Detection on Images from Satellite using Oriented Fast and Rotated Brief.
DOI: https://doi.org/10.36478/jeasci.2017.4500.4503
URL: https://www.makhillpublications.co/view-article/1816-949x/jeasci.2017.4500.4503