基于核拉普拉斯稀疏编码模型的图像分类.pdf

第卷第期模式识别与人工智能
27 年月10 Vol. 27 No. 10
2014 10 PR & AI Oct. 2014

基于核拉普拉斯稀疏编码模型的图像分类*
刘越1 彭宏京1 钱素静1 施炜2

1 南京工业大学电子与信息工程学院南京
( 211816)
2 南京机电职业技术学院计算机科学系南京
( 211135 )
摘要在稀疏词袋模型中由于码书的过完备性相似特征间稀疏编码存在多种组合方式从而导致相似的特征
, , ,
可能得到完全不同的编码文中提出基于核拉普拉斯稀疏编码的图像分类方法该方法首先通过在高维核空间中
. .
构造核拉普拉斯矩阵以描述特征间的几何依赖关系使相似特征的稀疏编码的相似性尽可能得到保持再采用交
, .
替固定码书与稀疏矩阵的方法优化目标函数进行码书学****并采用符号猜测法对特征进行稀疏编码最后用多类
, .
分类器分类实验证明文中方法可较大幅度降低量化误差提高分类准确率并在多个数据集上取得良好的测
SVM . , ,
试效果
.
关键词核方法拉普拉斯矩阵稀疏编码图像分类
, , ,
中图法分类号
TP 391. 4
Sparse Coding Model
Based on Kernel Laplacian for Image Classification
1 1 1 2
LIU Yue , PENG Hong-Jing , QIAN Su-Jing , SHI Wei
1 College of Electronics and Information Engineering Nanjing Tech University Nanjing
( , , 211816)
2 Department puter Science Nanjing Institute of Mechatronic Technology Nanjing
( , , 211135)
ABSTRACT
In bag-of-words with sparse coding model, similar features can be encoded as various sparse coding
combinations due to the pleteness of the codebook, which results in totally different visual
words. In this paper, a sparse coding method based on kernel Laplacian for image classification is
proposed. Firstly, a Laplacian matrix is constructed to capture geometric dependencies between the
features in high-dimensional kernel space, and thus the similarity of sparse coding between the similar
features can be