基于智能感知相似度测量的图像提取  

Image retrieval using intelligent similarity measure adapting to human perception

作　　者：肖建纳 腾师伟 阿利尔 张登胜 陆国军 刘颖[3] SHOJANAZERI Hamid;TENG Shyh Wei;ARYAL Sunil;ZHANG Dengsheng;LU Guojun;LIU Ying(School of Engineering,Information Technology and Physical Sciences,Federation University,VIC 3824,Australia;School of Info Technology,Deakin University,VIC 3125,Australia;Center for Image and Information Processing,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)

机构地区：[1]澳大利亚联邦大学工程信息和理工学院,维多利亚3842,澳大利亚 [2]迪肯大学信息学院,维多利亚3125,澳大利亚 [3]西安邮电大学图像信息处理中心,陕西西安710121

出　　处：《西安邮电大学学报》2021年第6期36-47,共12页Journal of Xi’an University of Posts and Telecommunications

基　　金：Australian Research Council Discovery Projects scheme(DP130100024);Service Enterprise Project of XI'AN Innovation Talents of Science and Technology(2020KJRC0110)。

摘　　要：相似度测量是图像分类和提取的重要内容。良好的相似度测量方法应该能以给定的图像特征类型从数据库中检索到相似图像,并在检索中丢弃不相关图像。基于距离的相似度测量,仅反映了高维特征空间中两个特征向量之间的空间距离,缺乏任何感知意义,而且还忽视了相似度决策过程中邻域的影响。新的感知相似度测量方法,既可以测量特征空间中两个图像间的距离,又能表达图像间的视觉相似性。实验数据表明,与常用的基于距离的相似度测量相比,新提出的相似度测量方法具有明显著优势。Similarity measure is an important component and research topic in image classification and retrieval.Given a type of image features,a good similarity measure should be able to retrieve similar images from the database while discard irrelevant images from the retrieval.Similarity measures in literature are typically distance based which measure the spatial distance between two feature vectors in high dimensional feature space.However,this type of similarity measures does not have any perceptual meaning and ignores the neighborhood influence in the similarity decision making process.In this paper,we propose a novel perceptual similarity measure which can measure both the distance and perceptual similarity of two image features in feature space.Our results show the proposed similarity measure has a significant improvement over the traditional distance based similarity measure most commonly used in literature.

关 键 词：图像提取 图像分类 相似度测量 依赖数据的相似度测量 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]