基于网格的图像内容检索研究
摘要
随着多媒体、网络技术的飞速发展,图像信息的应用日益广泛,如何从大量的图像信息中快速而准确地检索出所需要的信息,已成为一个亟待解决的问题。随着大规模图像数据库的产生,基于内容的图像检索技术成为图像领域研究的热点问题之一。在过去的几十年中,各地的研究者提出了各种不同的基于内容的图像检索方法,以提高图像检索的精度和速度,但是仍然不能满足人们的实际需要。网格的出现为解决这一问题提供了一条新的途径。
网格作为一种新型的分布式计算技术,逐渐成熟。网格具有超强的计算处理能力、良好的系统扩展能力和高效的分布式资源管理能力。它突破了计算能力的限制、存储能力的限制、资源分布的限制、资源共享方式的限制,适用于计算密集型、数据密集型应用。利用网格技术进行图像内容的检索是一种全新的尝试。为此,本文对基于网格的图像内容检索体系结构进行了研究,提出基于网格的图像检索的解决方案,即在网格中间层之上,为网格应用建立图像内容检索中间层,为用户提供了通用的图像检索平台。
本文首先对图像检索技术和网格技术作了介绍。然后,针对图像检索中存在的问题,提出了解决方案,把网格技术和图像检索相结合,并设计了一种基于网格的图像内容检索系统,详细阐述了图像检索网格系统的体系结构,以及对系统中的各个模块进行了详细的设计和规划,为解决当前图像检索问题提供了一种很有前景的解决方法和思路。本文还完成了Windows平台下GT4的安装、配置和测试工作,并构建了一个简单虚拟组织作为模拟网格环境,在模拟环境下,编写有关图像检索网格服务并对相关功能进行了测试。最后总结本文的研究工作,展望下一阶段的发展方向,为后续工作奠定了基础。
With the development of multimedia and Internet technology, the application of the image is more and more extensive. How to retrieve necessary information from large amount image information efficiently and quickly needs to be solved urgently. The Content-Based Image Retrieval (CBIR) has become one of the hot research areas in image domain for large image database. In the past several ten years, many researchers proposed various different CBIR algorithms to enhance the speed and accuracy of image retrieval, but that still can not meet the actual needs of people. The emergence of grid for the solution to this problem provides a new way.
As a new distributed technology, the grid technology has been mature. It provides an efficient managing way of distributed resource, strong computational power, excellent system expansibility, breaks the limit of computational power, storage, resource distributing, the way of sharing resource. It should be a fire-new experiment combining grid and CBIR technology. As a result, this paper researches on the architecture of CBIR based on grid, and proposes a solution of CBIR based on grid, namely, above the grid layer, a new layer of CBIR is build as a universal CBIR platform.
This paper first introduces the technology of CBIR and grid. Secondly, the paper puts forward a new scheme by combination of CBIR and grid in order to solve the resent problems of CBIR. Next, the paper designs a CBIR system based on grid and particularly expatiates on the framework of CBIR system and the design of each module of CBIR system. It provides a resolving method and idea with great prospect for CBIR. This Paper successfully installs, configuring and tests GT4 under the Windows operation system, establishes a simple virtual organization as a simulative Grid environment and procedures for the preparation of image retrieval in the simulative Grid environment to develop Grid services for testing related functions. Finally, this paper summarizes the writer's research work, looks forward to the developing directions and sets the base of succeeding work.
网格作为一种新型的分布式计算技术,逐渐成熟。网格具有超强的计算处理能力、良好的系统扩展能力和高效的分布式资源管理能力。它突破了计算能力的限制、存储能力的限制、资源分布的限制、资源共享方式的限制,适用于计算密集型、数据密集型应用。利用网格技术进行图像内容的检索是一种全新的尝试。为此,本文对基于网格的图像内容检索体系结构进行了研究,提出基于网格的图像检索的解决方案,即在网格中间层之上,为网格应用建立图像内容检索中间层,为用户提供了通用的图像检索平台。
本文首先对图像检索技术和网格技术作了介绍。然后,针对图像检索中存在的问题,提出了解决方案,把网格技术和图像检索相结合,并设计了一种基于网格的图像内容检索系统,详细阐述了图像检索网格系统的体系结构,以及对系统中的各个模块进行了详细的设计和规划,为解决当前图像检索问题提供了一种很有前景的解决方法和思路。本文还完成了Windows平台下GT4的安装、配置和测试工作,并构建了一个简单虚拟组织作为模拟网格环境,在模拟环境下,编写有关图像检索网格服务并对相关功能进行了测试。最后总结本文的研究工作,展望下一阶段的发展方向,为后续工作奠定了基础。
With the development of multimedia and Internet technology, the application of the image is more and more extensive. How to retrieve necessary information from large amount image information efficiently and quickly needs to be solved urgently. The Content-Based Image Retrieval (CBIR) has become one of the hot research areas in image domain for large image database. In the past several ten years, many researchers proposed various different CBIR algorithms to enhance the speed and accuracy of image retrieval, but that still can not meet the actual needs of people. The emergence of grid for the solution to this problem provides a new way.
As a new distributed technology, the grid technology has been mature. It provides an efficient managing way of distributed resource, strong computational power, excellent system expansibility, breaks the limit of computational power, storage, resource distributing, the way of sharing resource. It should be a fire-new experiment combining grid and CBIR technology. As a result, this paper researches on the architecture of CBIR based on grid, and proposes a solution of CBIR based on grid, namely, above the grid layer, a new layer of CBIR is build as a universal CBIR platform.
This paper first introduces the technology of CBIR and grid. Secondly, the paper puts forward a new scheme by combination of CBIR and grid in order to solve the resent problems of CBIR. Next, the paper designs a CBIR system based on grid and particularly expatiates on the framework of CBIR system and the design of each module of CBIR system. It provides a resolving method and idea with great prospect for CBIR. This Paper successfully installs, configuring and tests GT4 under the Windows operation system, establishes a simple virtual organization as a simulative Grid environment and procedures for the preparation of image retrieval in the simulative Grid environment to develop Grid services for testing related functions. Finally, this paper summarizes the writer's research work, looks forward to the developing directions and sets the base of succeeding work.
引文
[1]Gudivada V N,Raghavan V V.Content-based image retrieval systems[J].IEEE Computer,1995,28(9):18-22.
[2]王莲,张学福.基于内容的图像检索研究进展[J].现代情报,2005,25(5):25-31.
[3]Ian Foster,Carl Kesselman.网格计算(第二版)[M].北京:电子工业出版社,2004.
[4]W Niblack,et al.The QBIC project:Querying images by content using color,texture and shape[J].SPIE,1993,v1908:173-187.
[5]A Pentland,et al.Photobook:Tools for content-based manipulation of image databases[J],SPIE,1994,v2185:34-47.
[6]J R Smith,S F Chang.VisualSEEK:A Fully Automated Content-based Image Query System[C]//ACM Multimedia.Boston MA:MCM Press,1996,87-98.
[7]M Flickner,H Sawhney,et al.Query by image and video content:The QBIC system[J].IEEE Computer,1995,28(9):23-32.
[8]J Ashley,et al.Automatic and semi-automatic methods for image annotation and retrieval in QBIC[J].SPIE,1995,v2420,24-35.
[9]W Niblack,et al.Updates to the QBIC System[J].SPIE,1997,v3312,150-161.
[10]R.B.Jeffrey,et al.The Virage Image Search Engine:An Open Framework for Image Management[C]//Proc.SPIE Storage and Retrieval for Image and Video Database.Feb.1996.
[11]W.Y.Ma,B.S.Manjunath.Netra:A toolbox for navigating large image databases[J].Multimedia System,1999,7(3):184-198.
[12]T S Huang,S.Mehrotra,K Ramchandran.Multimedia Analysis and Retrieval System(MARS)Project[C]//Proc of 33 Rd Annual Clinic on Library Application of Data Processing-Digital Image Access and Retrieval,March 1996.
[13]金丰华.基于内容的医学图像检索[D].南京:东南大学,2003.
[14]朱兴全,张宏江.iFind:一个结合语义和视觉特征的图像相关反馈检索系统[J].计算机学报,2002,25(7):681-689.
[15]熊华,老松杨,吴玲琦.News Video CAR:一个基于内容的视频新闻节目浏览检索系统[J].计算机工程,2000,26(11):73-75.
[16]李鹏杰,杨树元.一种基于内容的图像检索系统ImageHunter[J].微计算机应用,2001,22(3):139-141.
[17]周丽华,周纪勤,柏正尧.基于图像信息熵的商标图像检索[J].计算机应用,2000,20(2):21-23.
[18]徐冰,王小华,顺国松.网格技术研发进展综述[J].杭州电子科技大学学报,2007,27(5):155-158.
[19]徐钊,刘志强.第三代Internet-网格及其关键技术研究[J].计算机与网络,2005,3:80.
[20]金海,郑纬民.网格计算专刊前言[J].软件学报,2006,17(11):2221-2223.
[21]中国科学院计算机网络信息中心:欧盟FP6计划BRIDGE项目启动[EB/OL].http://www.cnic.ac.cn/science/communion/200701230007.html.
[22]网格与服务计算研究中心[EB/OL].http://vega.ict.ac.cn/.
[23]秦金磊,朱有产,李玉凯.基于网格计算的关键技术研究[J].计算机技术与发展,2006,(6):7-8.
[24]I Foster,C Kesselman,S Tuecke.The anatomy of the grid:Enabling scalable virtual organizations.International Journal Supercomputer Applications,2001,15(3):200-222.
[25]夏靖波,刘颖,汪胜荣.网格原理与开发[M].西安:西安电子科技大学出版社,2006,5-10.
[26]都志辉,陈渝,刘鹏.网格计算[M].清华大学出版社,2002:9-12.
[27]桂小林.网格技术导论[M].北京:北京邮电大学出版社,2005.
[28]Ian Foster,Carl Kesselman.The Grid2:Blueprint for a New Computing Infrastructure[M].北京:电子工业出版社,2004.
[29]I Foster,C Kesselman.Grid Services for Distributed System Integration[J].Computer,2002,12(1):38-41.
[30]The WS-Resource Framework.http://www.globus.org/wsrf/.
[31]Karl Czajkowski,Donald F Ferguson,et al.The WS-Resource Framework Version 1.0,http://www.Globus.org/wsrf/specs/ws-wsrf.pdf.
[32]徐志伟,冯百明,李伟.网格计算技术[M].北京:电子工业出版社,2004,46-55.
[33]夏靖波,刘颖,汪胜荣.网格原理与开发[M].西安:西安电子科技大学出版社,2006.
[34]庄越挺,潘云鹤.基于内容的图像检索综述[J].摸式识别与人工智能,1999,12(2):170-177.
[35]Weber R.,Schek H.J,Blott S.A quantitative analysis and performance study for similarity-search methods in high-dimensional spaces[C]//In:Proc.24th Int.Conf.VLDB.New York.IEEE,1998,194-205.
[36]李俊山,李旭辉.数字图像处理[M].北京:消华大学出版社.2007.
[37]姚敏.数字图像处理[M].北京:机械工业出版社,2006.
[38]孙兴华.基于内容的图像检索研究[D].硕士论文,南京理工大学,2001.
[39]Swain M J,Ballard D H..Color indexing[J].International Journal of computer vision,1991,7(1):11-32.
[40]Stricker M,Orengo M.Similarity of color images[J].SPIE,1995,2420,381-392.
[41]Stricker M,Dimai A.Color indexing with weak spatial constraints[J].SPIE,1996,2670: 29-40.
[42]Kankanhalli M S,Mehtre B M.Cluster-based color matching for Image retrieval[J].Pattern Recognition,1996,29(4):701-709.
[43]孙延奎.小波分析及应用.北京:机械工业出版社,2005,100-116.
[44]李向阳,潘云鹤.基于椭圆的形状描述与检索方法[J].中国科学(E辑),2002,32(5):674-684.
[45]Aki.Kobayashi,Toshiyuki Yoshida,Sakai.Image Retrieval by Estimating Parameters of Distance Measure[J].SPIE,2000,3972:432-441.
[2]王莲,张学福.基于内容的图像检索研究进展[J].现代情报,2005,25(5):25-31.
[3]Ian Foster,Carl Kesselman.网格计算(第二版)[M].北京:电子工业出版社,2004.
[4]W Niblack,et al.The QBIC project:Querying images by content using color,texture and shape[J].SPIE,1993,v1908:173-187.
[5]A Pentland,et al.Photobook:Tools for content-based manipulation of image databases[J],SPIE,1994,v2185:34-47.
[6]J R Smith,S F Chang.VisualSEEK:A Fully Automated Content-based Image Query System[C]//ACM Multimedia.Boston MA:MCM Press,1996,87-98.
[7]M Flickner,H Sawhney,et al.Query by image and video content:The QBIC system[J].IEEE Computer,1995,28(9):23-32.
[8]J Ashley,et al.Automatic and semi-automatic methods for image annotation and retrieval in QBIC[J].SPIE,1995,v2420,24-35.
[9]W Niblack,et al.Updates to the QBIC System[J].SPIE,1997,v3312,150-161.
[10]R.B.Jeffrey,et al.The Virage Image Search Engine:An Open Framework for Image Management[C]//Proc.SPIE Storage and Retrieval for Image and Video Database.Feb.1996.
[11]W.Y.Ma,B.S.Manjunath.Netra:A toolbox for navigating large image databases[J].Multimedia System,1999,7(3):184-198.
[12]T S Huang,S.Mehrotra,K Ramchandran.Multimedia Analysis and Retrieval System(MARS)Project[C]//Proc of 33 Rd Annual Clinic on Library Application of Data Processing-Digital Image Access and Retrieval,March 1996.
[13]金丰华.基于内容的医学图像检索[D].南京:东南大学,2003.
[14]朱兴全,张宏江.iFind:一个结合语义和视觉特征的图像相关反馈检索系统[J].计算机学报,2002,25(7):681-689.
[15]熊华,老松杨,吴玲琦.News Video CAR:一个基于内容的视频新闻节目浏览检索系统[J].计算机工程,2000,26(11):73-75.
[16]李鹏杰,杨树元.一种基于内容的图像检索系统ImageHunter[J].微计算机应用,2001,22(3):139-141.
[17]周丽华,周纪勤,柏正尧.基于图像信息熵的商标图像检索[J].计算机应用,2000,20(2):21-23.
[18]徐冰,王小华,顺国松.网格技术研发进展综述[J].杭州电子科技大学学报,2007,27(5):155-158.
[19]徐钊,刘志强.第三代Internet-网格及其关键技术研究[J].计算机与网络,2005,3:80.
[20]金海,郑纬民.网格计算专刊前言[J].软件学报,2006,17(11):2221-2223.
[21]中国科学院计算机网络信息中心:欧盟FP6计划BRIDGE项目启动[EB/OL].http://www.cnic.ac.cn/science/communion/200701230007.html.
[22]网格与服务计算研究中心[EB/OL].http://vega.ict.ac.cn/.
[23]秦金磊,朱有产,李玉凯.基于网格计算的关键技术研究[J].计算机技术与发展,2006,(6):7-8.
[24]I Foster,C Kesselman,S Tuecke.The anatomy of the grid:Enabling scalable virtual organizations.International Journal Supercomputer Applications,2001,15(3):200-222.
[25]夏靖波,刘颖,汪胜荣.网格原理与开发[M].西安:西安电子科技大学出版社,2006,5-10.
[26]都志辉,陈渝,刘鹏.网格计算[M].清华大学出版社,2002:9-12.
[27]桂小林.网格技术导论[M].北京:北京邮电大学出版社,2005.
[28]Ian Foster,Carl Kesselman.The Grid2:Blueprint for a New Computing Infrastructure[M].北京:电子工业出版社,2004.
[29]I Foster,C Kesselman.Grid Services for Distributed System Integration[J].Computer,2002,12(1):38-41.
[30]The WS-Resource Framework.http://www.globus.org/wsrf/.
[31]Karl Czajkowski,Donald F Ferguson,et al.The WS-Resource Framework Version 1.0,http://www.Globus.org/wsrf/specs/ws-wsrf.pdf.
[32]徐志伟,冯百明,李伟.网格计算技术[M].北京:电子工业出版社,2004,46-55.
[33]夏靖波,刘颖,汪胜荣.网格原理与开发[M].西安:西安电子科技大学出版社,2006.
[34]庄越挺,潘云鹤.基于内容的图像检索综述[J].摸式识别与人工智能,1999,12(2):170-177.
[35]Weber R.,Schek H.J,Blott S.A quantitative analysis and performance study for similarity-search methods in high-dimensional spaces[C]//In:Proc.24th Int.Conf.VLDB.New York.IEEE,1998,194-205.
[36]李俊山,李旭辉.数字图像处理[M].北京:消华大学出版社.2007.
[37]姚敏.数字图像处理[M].北京:机械工业出版社,2006.
[38]孙兴华.基于内容的图像检索研究[D].硕士论文,南京理工大学,2001.
[39]Swain M J,Ballard D H..Color indexing[J].International Journal of computer vision,1991,7(1):11-32.
[40]Stricker M,Orengo M.Similarity of color images[J].SPIE,1995,2420,381-392.
[41]Stricker M,Dimai A.Color indexing with weak spatial constraints[J].SPIE,1996,2670: 29-40.
[42]Kankanhalli M S,Mehtre B M.Cluster-based color matching for Image retrieval[J].Pattern Recognition,1996,29(4):701-709.
[43]孙延奎.小波分析及应用.北京:机械工业出版社,2005,100-116.
[44]李向阳,潘云鹤.基于椭圆的形状描述与检索方法[J].中国科学(E辑),2002,32(5):674-684.
[45]Aki.Kobayashi,Toshiyuki Yoshida,Sakai.Image Retrieval by Estimating Parameters of Distance Measure[J].SPIE,2000,3972:432-441.