Detail Simulating traditional traffic data from satellite data: Preparing for real satellite data test Bibliografi Author: Jafar, Fatmeh Nazmi Ahmad ; McCord, Mark R. (Advisor); Merry, Carolyn J. (Advisor) Topik: ENGINEERING; CIVIL|REMOTE SENSING Bahasa: (EN )    ISBN: 0-599-66472-X     Penerbit: THE OHIO STATE UNIVERSITY     Tahun Terbit: 2000     Jenis: Theses - Dissertation Fulltext: 9962403.pdf (0.0B; 0 download) Abstract This research addresses demonstrating that vehicles can in fact be detected in satellite imagery. A possible test that would allow such a demonstration would compare traditional ground data to ground data simulated from satellite data. Ground data could be simulated from satellite imagery by estimating when the vehicles assumed to be observed in the satellite imagery would pass a ground counter location. Even if all vehicles could be observed in the imagery, there could still be errors when estimating the times the observed vehicles would pass the ground counter location. In this research, we investigate the errors involved with simulating ground data from satellite data and determine which errors seem most important. Specifically, we investigate distance and speed errors. We find that the speed errors have a much greater effect than the distance errors in estimating when imaged vehicles would pass the ground counter location and that the distance errors involved can essentially be ignored. We then investigate the effect of speed errors in estimating volume counts past a ground counter from imaged satellite data. We define a measure to reflect the variation in the volume counts estimated from the image-based volume estimates. The value of the variation measure is reduced when the time of the image acquisition is included in the time interval of the estimated count past the ground counter location. In this case, the variation does not depend on the length of the count interval and equals the difference between the ratios of maximum speed to average highway speed and minimum assumed speed to average highway speed. When simulating ground data from image data on a highway with a platoon of traffic concentrated at a specific segment of the highway, the count interval that includes the platoon of traffic can be determined. The variation in volume measure would be reduced for such a count interval. The variation in the volume measure is also shown to be reduced when the speeds of the vehicles at the time of the image are known. This effect is illustrated by an example using real image data. Opini Anda Klik untuk menuliskan opini Anda tentang koleksi ini! Lihat Sejarah Pengadaan  Konversi Metadata   Kembali

Copyright © 2006, 2007 Unika Atma Jaya, all rights reserved