Journal of Computer Applications ›› 2015, Vol. 35 ›› Issue (9): 2636-2639.DOI: 10.11772/j.issn.1001-9081.2015.09.2636

Previous Articles     Next Articles

Unsupervised deep learning method for color image recognition

KANG Xiaodong1, WANG Hao2, GUO Jun1, YU Wenyong1   

  1. 1. School of Medical Imaging, Tianjin Medical University, Tianjin 300070, China;
    2. Invasive Technology Department, Affiliated Hospital of Hebei University, Baoding Hebei 071000, China
  • Received:2015-04-30 Revised:2015-06-29 Online:2015-09-10 Published:2015-09-17


康晓东1, 王昊2, 郭军1, 于文勇1   

  1. 1. 天津医科大学 医学影像学院, 天津 300070;
    2. 河北大学附属医院 介入科, 河北 保定 071000
  • 通讯作者: 康晓东(1964-),男,天津人,教授,博士,CCF会员,主要研究方向:医学图像处理、医疗信息系统集成,
  • 作者简介:王昊(1984-),男,河北保定人,工程师,硕士,主要研究方向:图像处理;郭军(1972-),男,四川成都人,实验师,主要研究方向:实验技术;于文勇(1986-),男,山东青岛人,硕士研究生,主要研究方向:图像处理。

Abstract: In view of significance of color image recognition, the method of color image recognition based on data of image features and Deep Belief Network (DBN) was presented. Firstly, data field of color image was constructed in accord with human visual characteristics; secondly, wavelet transforms was applied to describe multi-scale feature of image; finally, image recognition could be made by training unsupervised DBN.The experimental results show that compared with the methods of Adaboost and Support Vector Machine(SVM),classification accuracy is improved by 3.7% and 2.8% respectively and better image recognition is achieved by the proposed method.

Key words: image recognition, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), computer vision

摘要: 针对彩色图像分类识别的重要性,提出了一种结合图像特征数据和深度信任网络(DBN)的彩色图像识别方法。首先,构造符合人类视觉特性的图像色彩数据场;其次,以小波变换描述图像的多尺度特征;最后,通过无监督训练深度信任网络实现对图像的识别。实验结果表明,所提方法与Adaboost、支持向量机(SVM)方法比较,分类准确率分别提高约3.7%和2.8%,可有效提高图像识别效果。

关键词: 图像识别, 深度信任网络, 受限玻尔兹曼机, 计算机视觉

CLC Number: