基于卷积神经网络和二进制K-means的图像快速聚类  

Fast Image Clustering Based on Convolutional Neural Network and Binary K-means

作　　者：柯圣财 李弼程 唐永旺[1] 吴志兵[3] 万建平 

机构地区：[1]解放军信息工程大学信息系统工程学院,郑州450002 [2]华桥大学计算机科学与技术学院,厦门361021 [3]江南计算技术研究所,无锡214083 [4]西安卫星测控研究中心,西安710000

出　　处：《数据采集与处理》2017年第5期970-979,共10页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(60872142)资助项目

摘　　要：当前主流的图像聚类方法采用的视觉特征缺乏自主学习能力,导致其图像表达能力不强,而且传统的聚类算法计算复杂度较高,聚类效率低,难以适应大数据环境。针对这些问题,本文提出了一种基于卷积神经网络和二进制K-means的图像快速聚类方法。首先,利用卷积神经网络学习图像内容的内在隐含关系,得到图像高阶特征,增强特征的视觉表达能力和区分性;然后,利用哈希方法将高维图像特征映射为低维二进制哈希码,并通过对聚类中心构造多索引哈希表来加速寻找最近的聚类中心,以降低时间复杂度;最后,利用二进制K-means完成二进制哈希码的快速聚类。在ImageNet-1000图像集上的实验结果表明,本文方法能够有效地增强图像特征的表达能力、提高图像聚类效率、性能优于当前主流方法。Visual features used in state-of-the-art image clustering methods lack of independent learning ability,which leads to low image expression ability.Furthermore,the efficiency of traditional clustering methods is low for large image dataset.So,a fast image clustering method based on convolutional neural network and binary K-means is proposed in this paper.Firstly,a large-scale convolutional neural network is employed to learn the intrinsic implications of training images so as to improve the discrimination and representational power of visual features.Secondly,hashing is applied to map high-dimensional deep features into low-dimensional hamming space,and multi-index hash table is used to index the initial centers so that the nearest center lookup becomes extremely efficient.Finally,image clustering is accomplished efficiently by binary K-means algorithm.Experimental results on ImageNet-1000 dataset indicate that the proposed method can effectively enhance the expression ability of image features,increase the image clustering efficiency and has better performance than state-of-the-art methods.

关 键 词：深度学习 图像聚类 卷积神经网络 二进制K-means 多索引哈希 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]