Title :
A geometric correction approach for chassis images with non-uniform speed by linear CCD camera
Author :
Cheng, Yan ; Liang, Lingfei ; Pu, Jiexin
Author_Institution :
Electron. & Inf. Eng., Henan Univ. of Sci. & Technol., Luoyang, China
Abstract :
This paper proposes a geometric correction method for vehicle chassis images acquired by linear charge coupled device (CCD) camera using speeded up robust features (SURF) and phase-only correlation (POC). We use feature matching based in SURF to estimate an approximate homography between standard images and real-time camera images using a limited number of corresponding points, and dense sub-pixel correspondence search based in POC to adjust precise position of corresponding points. Experimental result shows the geometric correction method using SURF and POC can achieve high-accuracy.
Keywords :
CCD image sensors; computational geometry; feature extraction; image matching; object detection; traffic engineering computing; POC; SURF; feature matching; geometric correction approach; linear CCD camera; linear charge coupled device camera; nonuniform speed; phase-only correlation; real-time camera images; speeded up robust features; vehicle chassis images; Cameras; Charge coupled devices; PSNR; Remote sensing; Robustness; Standards; Vehicles; POC; SURF; geometric correction; linear CCD camera;
Conference_Titel :
Automation and Logistics (ICAL), 2012 IEEE International Conference on
Conference_Location :
Zhengzhou
Print_ISBN :
978-1-4673-0362-0
Electronic_ISBN :
2161-8151
DOI :
10.1109/ICAL.2012.6308221