Title :
The two-dimensional code image tilt correction method based on least squares support vector machines
Author :
Yuanqian Cao ; Shicao Luo ; Yongsheng Ding ; Kuangrong Hao
Author_Institution :
Coll. of Inf. Sci. & Technol., Donghua Univ., Shanghai, China
Abstract :
In order to solve the two-dimensional code image tilt problem which affects locating the image, establishing and identifying sampling network, we suggest a two-dimensional image tilt correction method. In the proposed method, the least squares support vector machine (SVM) is used to regress the pixel coordinates on the two-dimensional code contour line, which can calculate the two-dimensional code image tilt vector and offset angle. This method converts the searching process for image tilt angle to solving linear matrix equation directly, which simplifies calculation and improves the efficiency of the algorithm. Also it avoids randomness and uncertainty in the process of searching. The simulation results show that this method can achieve good results for images that contain noise interference.
Keywords :
QR codes; image coding; image denoising; image sampling; least squares approximations; matrix algebra; support vector machines; SVM; image sampling; least square support vector machines; linear matrix equation; noise interference; offset angle; pixel coordinate regression; searching process; two-dimensional code contour line; two-dimensional code image tilt correction method; Algorithm design and analysis; Calibration; Educational institutions; Equations; Optimization; Support vector machines; Vectors; contour line extraction; least squares support vector machines; regression algorithm; the two-dimensional code; tilt correction;
Conference_Titel :
Audio, Language and Image Processing (ICALIP), 2014 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4799-3902-2
DOI :
10.1109/ICALIP.2014.7009930