Title :
Planar metric rectification from perspective image via vertical vanishing point
Author :
Tianyang Liu ; Weiming Xu ; Xiaodong Yin ; Xiliang Zhao
Author_Institution :
Dept. of Hydrogr. & Cartography, Dalian Naval Acad., Dalian, China
Abstract :
We propose a method for estimating the rectification homography of ground plane image using the plane´s vertical vanishing point. First, the method computes the image of the horizon by a vertical prism pole in the image. Then, the two imaged circular points of the ground plane are computed from the image of absolute conic and the imaged horizon. At last, the image rectification is computed from the conic dual to the circular points of the ground plane by singular value decomposition method. Experimental results show that the angle measurement error in the rectified image is less than 1.8°, which means the method is accurate and can be applied in various applications.
Keywords :
computational geometry; computer vision; singular value decomposition; absolute conic image; angle measurement error; ground plane image; imaged horizon; perspective image; planar metric rectification; rectification homography estimation; singular value decomposition method; vertical prism pole; vertical vanishing point; Calibration; Cameras; Equations; Image edge detection; Mathematical model; Noise level; circular points; metric rectification; normal line; vanishing point;
Conference_Titel :
Image and Signal Processing (CISP), 2014 7th International Congress on
Conference_Location :
Dalian
DOI :
10.1109/CISP.2014.7003831