Improved Diamond Search Algorithm

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 216)

Abstract

In order to improve the accuracy and speed of template matching, the paper presented a new fast search algorithm and called it the improved diamond search (IDS) algorithm. Through analyzing and carrying out several main search algorithms, the center-biased of motion field was taken into consideration in whole process. In addition, IDS algorithm added rough location and enhanced halfway-stop procedures to speed up convergence. Lots of vehicle tracking experiments show that the search performance of IDS algorithm is moderate. Moreover, contrastive experiments prove that IDS algorithm achieves faster search speed, supplying more effective technical support for template matching.

Keywords

Improved diamond search algorithm Motion vector Rough location 

Notes

Acknowledgments

This work was supported by the National Nature Science Foundation of China (61072148).

References

  1. 1.
    Liu W, Sun X, Yuan H (2011) Moving target tracking based on the improved diamond search algorithm and Gaussian scale-space. IHMSC 1:65–70Google Scholar
  2. 2.
    Hu W-C (2008) Adaptive template block-based block matching for object tracking. ISDA 1(3):61–64Google Scholar
  3. 3.
    Ni W, Guo B, Ding G (2005) A Hexagon-based motion estimation algorithm using motion vector field adaptive search technique. Comput Eng 31(13):Google Scholar
  4. 4.
    Kim JN, Choi TS (1998) A fast three-step search algorithm with minimum checking points using unimodal error surface assumption. IEEE Trans Consum Electron 44:638–648CrossRefGoogle Scholar
  5. 5.
    Li RX, Zeng B, Liou ML (1994) A new three-step search algorithm for block motion estimation. IEEE Trans Circuits Syst Video Technol 4:438–442CrossRefGoogle Scholar
  6. 6.
    Zhu S, Ma KK (2000) A new diamond search algorithm for fast block-matching motion estimation. IEEE Trans Image Proc 9:287–290CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  1. 1.Department of Information Science and TechnologyTaishan UniversityTaianChina

Personalised recommendations