Bag of Words - Quality Issues of Near-Duplicate Image Retrieval

Machine Graphics and Vision Pub Date : 2012-01-14 DOI:10.22630/mgv.2014.23.1.5

M. Paradowski, Mariusz Durak, Bartosz Broda

引用次数: 2

Abstract

This paper addresses the problem of large scale near-duplicate image retrieval. Issues related to visual words dictionary generation are discussed. A new spatial verification routine is proposed. It incorporates neighborhood consistency, term weighting and it is integrated into the Bhattacharyya coefficient. The proposed approach reaches almost 10\% higher retrieval quality, comparing to other recently reported state-of-the-art methods.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

词袋-近重复图像检索的质量问题

本文研究了大规模近重复图像检索问题。讨论了视觉词词典生成的相关问题。提出了一种新的空间验证程序。它结合了邻域一致性、项权重，并将其集成到Bhattacharyya系数中。与其他最近报道的最先进的方法相比，该方法的检索质量提高了近10%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Machine Graphics and Vision Computer Science-Computer Graphics and Computer-Aided Design

CiteScore

0.40

自引率

0.00%

发文量

期刊介绍： Machine GRAPHICS & VISION (MGV) is a refereed international journal, published quarterly, providing a scientific exchange forum and an authoritative source of information in the field of, in general, pictorial information exchange between computers and their environment, including applications of visual and graphical computer systems. The journal concentrates on theoretical and computational models underlying computer generated, analysed, or otherwise processed imagery, in particular: - image processing - scene analysis, modeling, and understanding - machine vision - pattern matching and pattern recognition - image synthesis, including three-dimensional imaging and solid modeling

期刊最新文献

Use of virtual reality to facilitate engineer training in the aerospace industry An efficient pedestrian attributes recognition system under challenging conditions Performance evaluation of Machine Learning models to predict heart attack Lung and colon cancer detection from CT images using Deep Learning Riesz-Laplace Wavelet Transform and PCNN Based Image Fusion