The Research and Design for Registering Detection of Rotary Screen Printing Machine Based on Mean Shift and Harris Operator

  • Junfeng Jing
  • Guanyan Li
  • Yang Li
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 113)


In order to realize the automatic registering system of rotary screen printing based on smart camera, this paper proposes the color printing images segmentation technology based on Mean Shift algorithm and the block matching algorithm based on Harris corner detection. The extended form of the Mean Shift algorithm is used in the segmentation of printing images, and the experiments show that the algorithm is good for the segmentation of printing images. Each chromaticity region is extracted from the segmented standard image,and the feature points on each region are detected by Harris operator. With these feature points as centers, select the standard matching blocks, and find the best matching blocks in the dealt image caught in real time to calculate the register error. The algorithms are proved to be feasible by simulations and experiments, which lays foundation for the closed-loop control of online register detection.


Rotary screen printing machine Register detection Mean shift Harris Block matching 



The authors would like to thank for the support of Scientific Research Program Funded by Shaanxi Provincial Education Department (Program No. 11JK0910).


  1. 1.
    Cheng Y (1995) Mean-shift mode seeking and clustering[J]. IEEE Trans Pattern Anal Machine Intell 17(8):790–799CrossRefGoogle Scholar
  2. 2.
    Fukunaga K, Hostetler LD (1975) The estimation of the gradient of adensity function, with applications in pattern recognition[J]. IEEE Trans Pattern Anal Machine Intell 21:32–40MATHMathSciNetGoogle Scholar
  3. 3.
    Xu M (2007) The study of a fabric image segmentation algorithm[D]. Zhejiang University, ZhejiangGoogle Scholar
  4. 4.
    Zhiqiang W, Zhixing C (2007) The convergence analysis of mean shift algorithm[J]. J Softw 182:205–212Google Scholar
  5. 5.
    Harris C, Satephens MJ (1988) A combined corner and edge detector[J]. Image Vis Comput 6(1):121–128Google Scholar
  6. 6.
    Lu H (2010) Research of block-matching motion estimation algorithm[D]. Jilin University, JilinGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  1. 1.College of Electronics and InformationXi’an Polytechnic UniversityXi’anChina
  2. 2.College of Mechanical and Electrical EngineeringXi’an University of Electronic Science and TechnologyXi’anChina

Personalised recommendations