An Improved Watershed Segmentation Algorithm for Adhesive Particles in Sugar Cane Crystallization

DOI

10.2991/icimm-15.2015.332How to use a DOI?

Keywords

Adhesive particles; Watershed; Image segmentation; Distance transformation; Area threshold; Shape factor

Abstract

Cane sugar crystallization, which has both heat and mass transfer, is a very complex chemical and physical process. Due to the constraints of technical conditions, detection of crystals is mainly relied on artificial methods with disadvantages of low accuracy and subjective judgment errors. Application of image processing technology can achieve obtaining the particle information of crystals automatically. However, adhesive phenomenon of collected image is serious, which causes difficulties for the subsequent image recognition. According to the problem of difficulties in segmenting adhesive particles for sugar cane crystallization, an improved watershed segmentation algorithm is proposed in this paper. Adhesive particles are remained by using area threshold and shape factor, and the single-completed particles are removed temporarily. The seed points are optimized by distance transformation, thus obtaining the simplified particle image. By using watershed algorithm to segment the simplified particle image and overlapping the single-completed particles with segmented result, the final image segmentation result is obtained. Experimental results show that the improved watershed algorithm can eliminate the irregular details, area noise and isolated edge pixels effectively.

Copyright

© 2015, 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  - Yanmei Meng
AU  - Kangyuan Zheng
AU  - Feihong Hu
AU  - Wenxing Li
AU  - Guancheng Lu
PY  - 2015/07
DA  - 2015/07
TI  - An Improved Watershed Segmentation Algorithm for Adhesive Particles in Sugar Cane Crystallization
BT  - Proceedings of the 5th International Conference on Information Engineering for Mechanics and Materials
PB  - Atlantis Press
SP  - 1774
EP  - 1779
SN  - 2352-5401
UR  - https://doi.org/10.2991/icimm-15.2015.332
DO  - 10.2991/icimm-15.2015.332
ID  - Meng2015/07
ER  -