Research on improved algorithm of Hoff circle detection

DOI

10.2991/wartia-16.2016.92How to use a DOI?

Keywords

Recognition accuracy, detect circles,Formation about Gradient.

Abstract

For the original random Hoff transform (RHT) in a small color image, the detection accuracy of the shortage of the circle, and if uncontrolled on the radius of the circle, not only the poor accuracy, but also a huge amount of computation.In order to improve the accuracy of finding circle, compared to the fitting, it need to provide additional parameters to control, parameter requirements more stringent, and poor overall stability. I proposed a modified RHT to find out the number of candidate circles after the vote, and then draw the circles, And comparison with the actual Contour,sorting by the actual total number of pixels overlap, then the highest score of the circle is the target circle.This method can solve the problem of low recognition accuracy. The following experiments we can see the error is improved, accuracy improvement has been very good. in improving the same time, it did not increase the amount of calculation. the algorithm has a high recognition accuracy, wide range, and strong anti-interference.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xing Zhang
AU  - Wei Lin
PY  - 2016/05
DA  - 2016/05
TI  - Research on improved algorithm of Hoff circle detection
BT  - Proceedings of the 2016 2nd Workshop on Advanced Research and Technology in Industry Applications
PB  - Atlantis Press
SP  - 460
EP  - 463
SN  - 2352-5401
UR  - https://doi.org/10.2991/wartia-16.2016.92
DO  - 10.2991/wartia-16.2016.92
ID  - Zhang2016/05
ER  -