Share Email Print

Proceedings Paper

An AK-LDMeans algorithm based on image clustering
Author(s): Huimin Chen; Xingwei Li; Yongbin Zhang; Nan Chen
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Clustering is an effective analytical technique for handling unmarked data for value mining. Its ultimate goal is to mark unclassified data quickly and correctly. We use the roadmap for the current image processing as the experimental background. In this paper, we propose an AK-LDMeans algorithm to automatically lock the K value by designing the Kcost fold line, and then use the long-distance high-density method to select the clustering centers to further replace the traditional initial clustering center selection method, which further improves the efficiency and accuracy of the traditional K-Means Algorithm. And the experimental results are compared with the current clustering algorithm and the results are obtained. The algorithm can provide effective reference value in the fields of image processing, machine vision and data mining.

Paper Details

Date Published: 5 March 2018
PDF: 9 pages
Proc. SPIE 10710, Young Scientists Forum 2017, 107101R (5 March 2018); doi: 10.1117/12.2317546
Show Author Affiliations
Huimin Chen, National Univ. of Defense Technology (China)
Xingwei Li, National Univ. of Defense Technology (China)
Yongbin Zhang, National Univ. of Defense Technology (China)
Nan Chen, National Univ. of Defense Technology (China)

Published in SPIE Proceedings Vol. 10710:
Young Scientists Forum 2017
Songlin Zhuang; Junhao Chu; Jian-Wei Pan, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?