DocumentCode :
3062922
Title :
The relation between accuracy and size of structure element for vehicle detection with high resolution highway aerial images
Author :
Shang Li ; Guoqing Zhou ; Zezhong Zheng ; Yalan Liu ; Xiaowen Li ; Ying Zhang ; Tao Yue
Author_Institution :
Sch. of Resources & Environ., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
fYear :
2013
fDate :
21-26 July 2013
Firstpage :
2645
Lastpage :
2648
Abstract :
It is a robust and efficient method that utilizes morphological operators, and Otsu threshold together with GIS road vector to detect vehicles from high resolution highway aerial images. Our previous experiments showed that the method resulted in a high correctness, completeness, and quality. However, we did not discuss the relation between the accuracy and the size of structure element for vehicle detection with high resolution highway aerial images. The space resolution of our aerial images is 0.15×0.15m, and most vehicles´ size is normally 5×2m, approaching 34×14 pixels. Thus, we selected a disc as structure element, but the disc´s radius should be less than 7 pixels. We compared the vehicle detection accuracy with 3 pixels, 4 pixels, 5 pixels, 6 pixels, and 7 pixels of disc with different size. Our result showed that the radius with 5 pixels or 6 pixels obtained the highest accuracy.
Keywords :
aerial equipment; vehicles; disc; high resolution highway aerial images; space resolution; structure element; vehicle detection; Accuracy; Gray-scale; Image resolution; Remote sensing; Road transportation; Vehicle detection; Vehicles; Vehicle detection; morphological operators; size of structure element;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2013 IEEE International
Conference_Location :
Melbourne, VIC
ISSN :
2153-6996
Print_ISBN :
978-1-4799-1114-1
Type :
conf
DOI :
10.1109/IGARSS.2013.6723366
Filename :
6723366
Link To Document :
بازگشت