Angle recognition algorithm based on discrete series data of object-boundary in polar coordinate

Abstract In computer vision applications on automation production lines, rotary angle of an object is often required to be calculated in the procedure of image processing. An angle recognition algorithm based on discrete array of object-boundary in polar coordinate is proposed. Firstly, calculate the object's center point and establish the polar coordinate using the object's center as the pole point. Then get the object's boundary through image processing method and discretize it according to the rule that they have the same central angles in the polar coordinate. Match the discretized array with that of the template image by step method and a rough rotary angle is obtained. Lastly, Liner interpolation method is used to get the exact solution. Practical matching experiments show that the method is simple and effective.

[1]  Sven Loncaric,et al.  A survey of shape analysis techniques , 1998, Pattern Recognit..

[2]  Wang Jun,et al.  Development and application of automation control system to plate production line , 2010, 2010 11th International Conference on Control Automation Robotics & Vision.

[3]  Marvin K. Simon,et al.  A simpler form of the Craig representation for the two-dimensional joint Gaussian Q-function , 2002, IEEE Communications Letters.

[4]  Qiuqi Ruan,et al.  Combining procrustes shape analysis and shape context descriptor for silhouette-based gait recognition , 2009 .

[5]  Shi Tie-ling.  Vision alignment system in automatic high precision chip mounter and its image processing , 2004 .

[6]  Ching-Long Shih,et al.  Locating and checking of BGA pins’ position using gray level , 2005 .

[7]  James B. Rawlings,et al.  Model-based object recognition to measure crystal size and shape distributions from in situ video images , 2007 .

[8]  Peter F. Stiller,et al.  Mathematical aspects of shape analysis for object recognition , 2007, Electronic Imaging.

[9]  Fatos T. Yarman-Vural,et al.  One-dimensional representation of two-dimensional information for HMM based handwriting recognition , 2000, Pattern Recognit. Lett..

[10]  Wang Guo.  Object Identification and Orientation Based on Contours , 2006 .

[11]  Jun Wang,et al.  A New High-Speed Algorithm for Center and Rotate Angle of Electronic Components , 2006, 2006 International Conference on Mechatronics and Automation.

[12]  Artyom M. Grigoryan,et al.  Fast Splitting$alpha$-Rooting Method of Image Enhancement: Tensor Representation , 2006, IEEE Transactions on Image Processing.