有监督子空间建模和稀疏表示的场景分类  

Scene categorization via supervised subspace modeling and sparse representation

作　　者：段菲[1] 章毓晋[1] 

机构地区：[1]清华大学电子工程系,北京100084

出　　处：《中国图象图形学报》2012年第11期1409-1417,共9页Journal of Image and Graphics

基　　金：国家自然科学基金项目(NNSF:61171118);教育部高等学校博士学科点专项科研基金项目(SRFDP-20110002110057)

摘　　要：提出了一种基于有监督子空间建模和稀疏表示的场景分类算法。该算法将采用非监督方式求取所有场景类别公共字典的稀疏编码模型分解为一系列各目标函数相互独立的多目标优化问题,实现了各类别字典的有监督学习。在所有类别的字典学习完毕后,再以各子空间和的基集来对每幅图像中所有局部特征进行协同编码,并借助空间金字塔表示(SPR)和特征各维最大汇总(max pooling)构成最终图像的全局特征表示。为对算法的有效性进行验证,在4个常用的场景图像库上进行了分类实验,结果表明该算法比采用非监督字典学习的方法在性能上有了显著提升。In this paper, we present a new scene categorization algorithm based on supervised subspace modeling and sparse representation. The proposed method implements supervised dictionary learning via decomposing the unsupervised sparse dictionary learning model into a group of independent optimization problems. After learning the dictionaries of all cat- egories, we aggregate them to form a global dictionary and encode each local feature of an image based on it. After using spatial pyramid representation and max pooling of local features＇ coding vectors, the final holistic feature depicting a scene image can be retrived. Comprehensive experimental results on four popular benchmark scene datasets show that our method achieves very promising result compared to existing state-of-the-art techniques.

关 键 词：稀疏表示 字典学习 空间金字塔匹配 场景分类 子空间建模 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]