Share Email Print

Proceedings Paper

The research of road and vehicle information extraction algorithm based on high resolution remote sensing image
Author(s): Tingting Zhou; Lingjia Gu; Ruizhi Ren; Qiong Cao
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

With the rapid development of remote sensing technology, the spatial resolution and temporal resolution of satellite imagery also have a huge increase. Meanwhile, High-spatial-resolution images are becoming increasingly popular for commercial applications. The remote sensing image technology has broad application prospects in intelligent traffic. Compared with traditional traffic information collection methods, vehicle information extraction using high-resolution remote sensing image has the advantages of high resolution and wide coverage. This has great guiding significance to urban planning, transportation management, travel route choice and so on. Firstly, this paper preprocessed the acquired high-resolution multi-spectral and panchromatic remote sensing images. After that, on the one hand, in order to get the optimal thresholding for image segmentation, histogram equalization and linear enhancement technologies were applied into the preprocessing results. On the other hand, considering distribution characteristics of road, the normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used to suppress water and vegetation information of preprocessing results. Then, the above two processing result were combined. Finally, the geometric characteristics were used to completed road information extraction. The road vector extracted was used to limit the target vehicle area. Target vehicle extraction was divided into bright vehicles extraction and dark vehicles extraction. Eventually, the extraction results of the two kinds of vehicles were combined to get the final results. The experiment results demonstrated that the proposed algorithm has a high precision for the vehicle information extraction for different high resolution remote sensing images. Among these results, the average fault detection rate was about 5.36%, the average residual rate was about 13.60% and the average accuracy was approximately 91.26%.

Paper Details

Date Published: 19 September 2016
PDF: 13 pages
Proc. SPIE 9977, Remote Sensing System Engineering VI, 99770H (19 September 2016); doi: 10.1117/12.2235337
Show Author Affiliations
Tingting Zhou, Jilin Univ. (China)
Lingjia Gu, Jilin Univ. (China)
Ruizhi Ren, Jilin Univ. (China)
Qiong Cao, Jilin Univ. (China)

Published in SPIE Proceedings Vol. 9977:
Remote Sensing System Engineering VI
Philip E. Ardanuy; Jeffery J. Puschell, Editor(s)

© SPIE. Terms of Use
Back to Top