A massive images classification method based on MapReduce parallel fuzzy C-means clustering

Article type: Research Article

Authors: Hu, Jinping | Cheng, Qian | Wen, Zhicheng^*

Affiliations: Jiangxi University of Engineering, Xinyu, Jiangxi, China

Correspondence: [*] Corresponding author: Zhicheng Wen, Jiangxi University of Engineering, Xinyu, Jiangxi, China. Tel.: +86 15873372746; E-mail: zcwen@mail.shu.edu.cn.

Abstract: Aiming at the low performance of classifying images under the computing model of single node. With GLCM (Gray Level Co-occurrence Matrix) which fuses gray level with texture of image, a parallel fuzzy C-means clustering method based on MapReduce is designed to classify massive images and improve the real-time performance of classification. The experimental results show that the speedup ratio of this method is more than 10% higher than that of the other two methods, moreover, the accuracy of image classification has not decreased. It shows that this method has high real-time processing efficiency in massive images classification.

Keywords: Gray level-texture Co-occurrence Matrix, MapReduce, fuzzy C-means, massive images classification

DOI: 10.3233/JCM-204829

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 21, no. 4, pp. 999-1011, 2021