计算机应用 ›› 2018, Vol. 38 ›› Issue (1): 233-237.DOI: 10.11772/j.issn.1001-9081.2017061394

• 虚拟现实与多媒体计算 • 上一篇    下一篇

具有普适性的改进非负矩阵分解图像特征提取方法

贾旭1, 孙福明1, 李豪杰2, 曹玉东1   

  1. 1. 辽宁工业大学 电子与信息工程学院, 辽宁 锦州 121001;
    2. 大连理工大学 软件学院, 辽宁 大连 116024
  • 收稿日期:2017-06-07 修回日期:2017-08-05 出版日期:2018-01-10 发布日期:2018-01-22
  • 通讯作者: 贾旭
  • 作者简介:贾旭(1983-),男,辽宁开原人,副教授,博士,CCF会员,主要研究方向:模式识别、机器学习;孙福明(1972-),男,辽宁大连人,教授,博士,主要研究方向:多媒体处理、机器学习;李豪杰(1973-),男,辽宁大连人,教授,博士,主要研究方向:多媒体处理、计算机视觉;曹玉东(1971-),男,辽宁昌图人,副教授,博士,主要研究方向:模式识别、图像处理。
  • 基金资助:
    国家自然科学基金资助项目(61502216,61572244)。

Image feature extraction method based on improved nonnegative matrix factorization with universality

JIA Xu1, SUN Fuming1, LI Haojie2, CAO Yudong1   

  1. 1. School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou Liaoning 121001, China;
    2. School of Software Technology, Dalian University of Technology, Dalian Liaoning 116024, China
  • Received:2017-06-07 Revised:2017-08-05 Online:2018-01-10 Published:2018-01-22
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61502216, 61572244).

摘要: 为提高图像特征提取的普适性,提出了一种基于改进非负矩阵分解(NMF)的图像特征提取方法。首先,考虑到提取的图像特征的实际意义,选用非负矩阵分解模型进行图像特征的降维处理;其次,为实现用较小数量系数来描述图像特征,将稀疏约束作为非负矩阵分解模型的正则项之一;然后,为使降维后优化得到的特征具有较好的类间区分性,将聚类属性作为非负矩阵分解的另一个正则项;最后,通过对模型的梯度下降优化求解,获得最优的特征基向量与图像特征向量。实验结果表明,针对3种图像数据库,所提的图像特征更有利于图像正确分类或识别,错误接受率(FAR)与错误拒绝率(FRR)分别可以降低到0.021与0.025。

关键词: 非负矩阵分解, 特征提取, 稀疏表示, 梯度下降法, 特征降维

Abstract: To improve the universality of image feature extraction, an image feature extraction method based on improved Nonnegative Matrix Factorization (NMF) was proposed. Firstly, considering the practical significance of extracted image features, NMF model was used to reduce the dimension of image feature vector. Secondly, in order to represent the image by a small number of coefficients, a sparse constraint was added to the NMF model as one of the regular terms. Then, to make the optimized feature have better inter-class differentiation, the clustering property constraint would be another regular term of the NMF model. Finally, through optimizing the model by using gradient descent method, the best feature basis vector and image feature vector could be acquired. The experimental results show that for three image databases, the acquired features extracted by the improved NMF model are more conducive to correct image classification or identification, and the False Accept Rate (FAR) and False Reject Rate (FRR) are reduced to 0.021 and 0.025 respectively.

Key words: Nonnegative Matrix Factorization (NMF), feature extraction, sparse representation, gradient descent method, feature dimension reduction

中图分类号: