一种基于词袋模型的大规模图像层次化分组算法 |
| |
引用本文: | 钱钧,杨恒,刘培桢,姜文涛,周锋飞. 一种基于词袋模型的大规模图像层次化分组算法[J]. 应用光学, 2014, 35(5): 799-805 |
| |
作者姓名: | 钱钧 杨恒 刘培桢 姜文涛 周锋飞 |
| |
作者单位: | 1.西安应用光学研究所,陕西 西安 710065 |
| |
摘 要: | 大规模图像集合的自动分组,不仅可以帮助用户快速组织和掌握图像集合的内容,并且是基于图像的三维场景重建应用的前提和重要环节。提出一种基于词袋模型(bag-of-words, BOW)的层次化分组算法,将每幅图像表示为一个超高维视词向量,利用多路量化技术将内容相似的图像量化到同一个节点,从而完成对图像粗略分组。然后,在每组类别里面,对图像的局部特征向量进行逐一匹配,并利用仿射空间不变量的约束条件,去除不可靠特征匹配,得到更为准确可靠的图像相似度度量,从而完成图像的精细分组。实验结果表明:从得到的系统不同阶段图像分组的查准率-查全率(precision-recall)曲线可以看出,精细分组过程可以显著提高粗分组精度,并且在精细分组阶段,使用约束条件比不使用约束还能获得更高的分组精度
|
关 键 词: | 图像分组 词袋模型 多路量化 仿射不变量约束 特征匹配 |
收稿时间: | 2014-04-16 |
Large-scale image hierarchical grouping algorithm based on bag-of-words |
| |
Affiliation: | 1.Xi-an Institute of Applied Optics,Xi-an 710065,China |
| |
Abstract: | Automatical grouping algorithm on large-scale image set ,which is an important part of the scene reconstruction system, can help users organize the image set contents quickly. A hierarchical image grouping algorithm based on bag-of-words(BOW) was proposed. Firstly, each image is projected to a super high dimensional visual word vector by a multiple paths quantization (MPQ) method, and this step is so-called coarse grouping. Then, feature matching is carried out in every divided group and an affine invariant constraint is proposed to get rid of the incorrect matching features. This step is so-called refined grouping which can improve image grouping accuracy. The precision-recall curves show that the refined grouping can obviously improve the accucy of coase grouping ,and better grouping accucy can be achieved when using constraints. |
| |
Keywords: | image grouping bag-of-words(BOW) multiple-path quantization affine invariant constrains feature matching |
|
| 点击此处可从《应用光学》浏览原始摘要信息 |
|
点击此处可从《应用光学》下载免费的PDF全文 |
|