Improved SLIC Superpixel Segmentation Based on HSV Non-uniform Quantization

DOI

10.2991/icimm-16.2016.52How to use a DOI?

Keywords

SLIC, superpixel, segmentation, HSV, quantization

Abstract

Simple linear iterative clustering (SLIC) method needs to artificially set the number of superpixel required for segmentation, which will easily lead to over-segmentation or under-segmentation, and that results can not meet the visual requirements. To improve the SLIC superpixel segmentation method, HSV non-uniform quantization, on the basis of color information,is proposed to get the picture feature vector for visual perception oriented, and then get the superpixel numbers of images needed. Experiments show that, over the number of superpixel obtained by this method, images can be partitioned into more desirably regions.

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  - Dongping Li
AU  - Changliang Liu
PY  - 2016/11
DA  - 2016/11
TI  - Improved SLIC Superpixel Segmentation Based on HSV Non-uniform Quantization
BT  - Proceedings of the 6th International Conference on Information Engineering for Mechanics and Materials
PB  - Atlantis Press
SP  - 277
EP  - 281
SN  - 2352-5401
UR  - https://doi.org/10.2991/icimm-16.2016.52
DO  - 10.2991/icimm-16.2016.52
ID  - Li2016/11
ER  -