计算机应用 ›› 2012, Vol. 32 ›› Issue (12): 3315-3318.DOI: 10.3724/SP.J.1087.2012.03315

• 人工智能 • 上一篇    下一篇

成对约束指导的稀疏保持投影

齐鸣鸣1,2   

  1. 1. 绍兴文理学院 元培学院,浙江 绍兴 312000
    2. 同济大学 计算机科学与技术系,上海 20180
  • 收稿日期:2012-08-01 修回日期:2012-09-06 发布日期:2012-12-29 出版日期:2012-12-01
  • 通讯作者: 齐鸣鸣
  • 作者简介:齐鸣鸣(1974-),男,江西景德镇人,讲师,博士研究生,主要研究方向:机器学习、图像处理。
  • 基金资助:
    国家自然科学基金资助项目

Pairwise constraint-guided sparsity preserving projections

QI Ming-ming1,2   

  1. 1. Department of Computer Science and Technology, Tongji University, Shanghai 201804,China
    2. School of Yuanpei, Shaoxing University, Shaoxing Zhejiang 312000,China
  • Received:2012-08-01 Revised:2012-09-06 Online:2012-12-29 Published:2012-12-01
  • Contact: QI Ming-ming

摘要: 针对稀疏保持投影的稀疏重构过程中监督信息不足的问题,提出一种成对约束指导的稀疏保持投影算法。该算法在训练样本数据的稀疏重构的过程中,通过引入正约束和负约束监督信息指导稀疏重构,使得稀疏保持投影有效地融合了约束监督信息。在UMIST、YALE和AR人脸库人脸数据集上的实验结果表明,与无监督的稀疏保持投影相比,该方法提高了基于最近近邻分类算法的5%~15%识别准确率,有效地提高了降维分类性能。

关键词: 降维, 稀疏重构, 成对约束, 稀疏保持投影

Abstract: Concerning the deficiency of supervision information in the process of sparse reconstruction in Sparsity Preserving Projection (SPP), Pairwise Constraint-guided Sparsity Preserving Projection (PCSPP) was proposed, which introduced supervision information of must-link constraints and cannot-link constraints to guide sparse reconstruction in the process of sparsity reconstruction of training samples, making SPP fuse constraint supervise information efficiently. The experimental results in UMIST,YALE and AR face datasets show, in contrast to unsupervised sparsity preserving projections, our algorithm achieves approximately 5%~15% increase in recognition accuracy based on the nearest neighbor classifier and promotes efficiently the performance of dimensionality reduction classification.

Key words: Dimensionality reduction, Sparse reconstruction, Pairwise constraint, Sparsity preserving projections

中图分类号: