DocumentCode
3013429
Title
High-speed distance measurement between moving vehicles with NIR-LED markers
Author
Fukuzawa, M. ; Hama, H. ; Nakamori, N. ; Yamada, M.
Author_Institution
Graduation Sch. of Sci. & Technol., Kyoto Inst. of Technol., Kyoto
fYear
2008
fDate
24-27 Dec. 2008
Firstpage
516
Lastpage
520
Abstract
An image-based technique with high measurement rate has been proposed to measure the distance between two moving vehicles. By combining several near-infrared (NIR) LED markers mounted on the forward vehicle with an NIR-filtered CMOS camera mounted on the backward vehicle, one can clearly recognize the marker shape and positions without any adjusting the focus lens of camera, and hence he can extremely reduce the time of digital preprocessing such as noise elimination and marker segmentation. The distance may be easily calculated from the geometrical transformation of the marker positions obtained in sub-pixel resolution from the gravity center of the pixel value. The image-based technique proposed here is actually examined in the vehicles moving on linear and curved tracks. It is experimentally confirmed that the measurement rate attains to 1000 frames/sec in the windowed image of 752 times 34 pixels, with a small processor with the clock of 600 MHz, including the capture time of the image, the calculation time of the gravity center, and the geometrical transformation time. The measurement error is less than 1 % for the distance range of 0.5 to 3 m.
Keywords
CMOS image sensors; cameras; computational geometry; distance measurement; filtering theory; image motion analysis; image resolution; infrared imaging; light emitting diodes; object detection; road vehicles; shape recognition; traffic engineering computing; NIR-LED marker; NIR-filtered CMOS camera; geometrical transformation; high-speed distance measurement; marker shape recognition; moving vehicle detection; near-infrared light emitting diode; sub-pixel image resolution; sub-pixel resolution; CMOS image sensors; Digital cameras; Distance measurement; Focusing; Gravity; Lenses; Light emitting diodes; Noise reduction; Shape; Vehicles; distance measurement; image processing; infrared imaging; motion control; moving vehicles;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer and Information Technology, 2008. ICCIT 2008. 11th International Conference on
Conference_Location
Khulna
Print_ISBN
978-1-4244-2135-0
Electronic_ISBN
978-1-4244-2136-7
Type
conf
DOI
10.1109/ICCITECHN.2008.4803048
Filename
4803048
Link To Document