基于内容的图像检索
|
摘要
基于内容的图像检索是多媒体信息处理中的一个重要问题。图像、视频作为多媒体中最直观、最形象的内容,对它们的检索和查询是多媒体信息处理的一个重要方面。本文首先回顾了基于内容的图像检索理论及其应用研究的整个发展过程,全面综述了在图像检索,尤其是基于内容的图像检索领域的技术和现状,并探讨了图像检索中的一些关键技术。
纹理是图像最重要的特征之一,在很多基于内容的图像检索中都作为重要的特征用来匹配与查询。本文首先使用基于共生矩阵的纹理特征进行检索,为了提高检索准确率,提出综合纹理和像素中心的检索方法。对花卉、植物的图像库的实验表明,该方法比使用单一纹理特征检索的效果要好。
压缩域的图像检索是基于内容的图像检索的一个热点,本文在这方面也作了一定的探讨和研究,提出了使用拼贴误差直方图的分形图像检索方法。拼贴误差是值域块和“最匹配的”定义域块相似性的一个度量,且直方图使用统计特征,具有计算简单的特点。所以该方法不仅有一定的检索准确性,最重要的是大大减少了计算复杂度。实验结果证明,该方法的确既能使计算简便,又能保证检索的准确性。
Content Based Image Retrieval (CBIR) is an important issue in multimedia information processing. As the most direct and vivid content, retrieval of image and video information is an important aspect in multimedia information processing.
In this thesis, firstly a brief review of the theory, applications and development of CBIR are presented, after which, the techniques and the current state of image retrieval, especially the state of CBIR are summarized. Finally, some important techniques in Content Based Image Retrieval System (CBIRS) are introduced.
Texture is one of the most important features of an image. Almost all of the presented CBIR systems use texture as an essential feature for matching and retrieval. In this thesis, a method utilizing both texture features and image center is proposed for retrieval. Experimental results on image database of real flowers and plants show that this method out-performs the method only with texture features based on the co-occurrence matrix.
At the same time, CBIR techniques in the compressed domain have attracted much interest. A new fractal coding based indexing technique with histogram of collage errors as the retrieval keys is proposed. Collage error is a quantitative measure of the similarity between the range block and the
“best-matched”domain block. Meanwhile, histograms can capture statistical characteristics and can be easily computed. So the proposed method can not only reduce computational complexities greatly, but also enhance the retrieval accuracy. Experimental results on a database of more than 200 texture images illustrate that the proposed method performs excellently.
纹理是图像最重要的特征之一,在很多基于内容的图像检索中都作为重要的特征用来匹配与查询。本文首先使用基于共生矩阵的纹理特征进行检索,为了提高检索准确率,提出综合纹理和像素中心的检索方法。对花卉、植物的图像库的实验表明,该方法比使用单一纹理特征检索的效果要好。
压缩域的图像检索是基于内容的图像检索的一个热点,本文在这方面也作了一定的探讨和研究,提出了使用拼贴误差直方图的分形图像检索方法。拼贴误差是值域块和“最匹配的”定义域块相似性的一个度量,且直方图使用统计特征,具有计算简单的特点。所以该方法不仅有一定的检索准确性,最重要的是大大减少了计算复杂度。实验结果证明,该方法的确既能使计算简便,又能保证检索的准确性。
Content Based Image Retrieval (CBIR) is an important issue in multimedia information processing. As the most direct and vivid content, retrieval of image and video information is an important aspect in multimedia information processing.
In this thesis, firstly a brief review of the theory, applications and development of CBIR are presented, after which, the techniques and the current state of image retrieval, especially the state of CBIR are summarized. Finally, some important techniques in Content Based Image Retrieval System (CBIRS) are introduced.
Texture is one of the most important features of an image. Almost all of the presented CBIR systems use texture as an essential feature for matching and retrieval. In this thesis, a method utilizing both texture features and image center is proposed for retrieval. Experimental results on image database of real flowers and plants show that this method out-performs the method only with texture features based on the co-occurrence matrix.
At the same time, CBIR techniques in the compressed domain have attracted much interest. A new fractal coding based indexing technique with histogram of collage errors as the retrieval keys is proposed. Collage error is a quantitative measure of the similarity between the range block and the
“best-matched”domain block. Meanwhile, histograms can capture statistical characteristics and can be easily computed. So the proposed method can not only reduce computational complexities greatly, but also enhance the retrieval accuracy. Experimental results on a database of more than 200 texture images illustrate that the proposed method performs excellently.
引文
[1]Smeulders A W M,Worring M,Santini S,et al.Content-based imageretrieval at the end of the early years[J].IEEE transactions on patternanalysis and machine intelligence.2000.22(12):1349-1379.
[2]章毓晋.国际标准MPEG-7问答.中国图像图形学报.2000.5B(11-12):7-10.
[3]Faloutsos C et al.Efficient and Effective Querying by Image Content.Research Report RJ9453(B3704).1993.IBM Research Division.SanJose,Califor J Intell Information Systems.1994.3:231-262.
[4]Flickner M et al.Querying by Image and Video Content:The QBICSystem.Computer.1995:23-32.
[5]Niblack W et al.The QBIC Project:Querying Images by Content UsingColor,Texture and Shape.1993.SPIE,1908:173-187.
[6]庄越挺.智能多媒体信息分析与检索的研究[D].浙江大学博士论文.1998.
[7]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,(5):139-141.
[8]王绍棣.基于内容特征的图像和视频检索系统[J].计算机工程与应用.2001,(7):113-117.
[9]Greg Pass,Ramin Zabih,and Justin Miller.Comparing images usingcolor coherence vectors.Proc.ACM Conf.On Multimedia.1996.
[10]Huang J,Kumar S,Mitra M,Zhu W J,and Zabih R.Image indexingusing color correlogram.Proc.of IEEE Conf.On Computer Vision andPattern Recognition.1997.
[11]Smith J R and Chang S F.Local color and texture extraction and spatialquery.Proc.IEEE Int.Conf.on Image Proc.1996
[12]Ma W Y,Deng Yining,and Manjunath B.S..Tools for texture/colorbased search of images.SPIE Int.Conf.3106,Human Vision andElectronic Imaging.Feb 1997.
[13]Haralick Robert M,Shanmugam K.Texture Features for ImageClassification.IEEE Trans.on Systems.Man,Cybernetics.1973,SMC-3(6):610-621.
[14]Tamura Hideyuki,Mori Shunji,and Yamawaki Takashi.Texture featurescorresponding to visual perception.IEEE Trans.on Sys,Man,andCyb.1978.SMC-8(6).
[15]Ma W.Y.and Manjunath B.S..A comparison of wavelet transformfeatures for texture image annotation.Proc.IEEE Int.Conf.on Image.1994.
[16]Pentland A,Picard R.W.and Sclaroff S.Photobook:Content-basedmanipulation of image databases.International Journal of ComputerVision.1996.
[17]Chuang Gene C.H.and Kuo C.C.Jay.Wavelet descriptor of planar curves:Theory and applications.IEEE Trans.Image Proc..January,1996,5(1):56-70.
[18]Kass M,Witkin A,Terzopoulos D.Snake:Active Contour Model.IJCV,1988,1(4):321-331.
[19]Rickman Rick and Stonham John.Content-based image retrieval usingcolor tuple histograms.Proc.SPIE Storage and Retrieval for Image andVideo Databases.1996.
[20]Stone H.S.,Li C.S.,Image matching by means of intensity and texturematching in the Fourier domain.Proc.SPIE.1996,Vol.2670:337-349.
[21]Celentano A.and Lecce V.D..A FFT based technique for imagesignature generation.Proc.of SPIE:Storage and Retrieval for Image andVideo Databases V.Feb 1997,Vol.3022:457-466.
[22]Smith J.R.and S.F.Transform features for texture classification anddiscrimination in large image databases.Proc of IEEE Intl.Conf on ImageProcessing.1994.Vol.3:407-411.
[23]Reeves R.,Kubik K.and Osberger W..Texture characterization ofcompressed aerial images using DCT coefficients.Proc.of SPIE:Storageand Retrieval for Image and Video DatabasesV.Feb,1997.Vol.3022:398-407,
[24]Shneier M.and Motaleb M.A..Exploiting the JPEG compression schemefor image retrieval.IEEE Trans on Pattern Analysis and MachineIntelligence.Aug 1996.Vol.18,No.8:849-853.
[25]Chang T.and Kuo C.C.J..Texture analysis and classification withtree-structured wavelet transform.IEEE Trans.on Image Processing.Oct1993.Vol.2,No.4:429-441.
[26]Manjunath B.S.and Ma W.Y..Texture features for browsing and retrieval of image data.IEEE Trans.On Pattern Analysis and MachineIntelligence.Aug 1996.Vol.18,No.8:837-841.
[27]Muneesawang P,Guan L.Multiresolution histogram indexing andrelevance feed back learning for image retrieval.In:Proceedings of IEEEon Image Processing Vancouver,BC,Canada.2000:526.
[28]Manjunath B.S.and Ma W.Y..Image Indexing Using a TextureDictionary.Proc.of SPIE Conf.on Image Storage and Archiving System,volume 2606.
[29]Amos Tversky.Features of similarity.Psychological review.1977.84(4):327-352.
[30]Simone Santini and Ramesh Jain.Similarity Match.IEEE Trans.onPattern Analysis and Machine Intelligence.1996.
[31]Furht B et al.Video and Image Processing in Multimedia Systems.Boston:Kluwer Academic Publishers.1995:226-270.
[32]Perry B,et al.Content-based Access to Multimedia Information.KluwerAcademic Publishers.1999.
[33]崔锦泰,程正兴,白居宪.小波分析导论.西安交通大学出版社.1995.
[34]Aksoy S,Haralick R M.Feature normalization and likelihood-basedsimilarity measures for image retrieval.PRL,2001,22(5):563-582.
[35]Hemami S S.Image compression-a review.In:Image Databases-Searchand Retrieval of Digital Imagery.Castelli V,Bergman L D,eds.JohnWiley&Sons,Inc.2002.Ch.8:211-239.
[36]Jacquin A E.Fractal image coding:A review.Proc.IEEE.1993.81(10):1451-1465.
[37]曾文曲,文有为,孙炜.分形小波与图像压缩.沈阳:东北大学出版社.2002年第一版:第二章.
[38]Sloan A D.Retrieving database contents by image recognition:Newfractal power.Advanced Imaging.1994.9(5):26-30.
[39]Zhang A,Cheng B,Acharya R.An approach to query-by-texture inimage database systems.In Proceedings of the SPIE Conference onDigital Image Storage and Archiving Systems,Philadelphia,USA.Oct,1995:338-349.
[40]Zhang Aidong,Cheng Biao,Acharya Raj,et al..Comparison of wavelettransforms and fractal coding in texture based image retrieval.InProceedings of the SPIE Conference on Visual Data Exploration andAnalysisⅢ,San Jose.Jan,1996:116-125.
[41]Schouten B,Zeeuw P.Image databases,scale and fractal transforms.Proceeding of ICIP,Canada.Sept 2000,vol.2:534-537.
[42]Pi M,Mandal M,Basu A.Image retrieval based on histogram of newfractal parameters.Proceeding of ICASSP,Hong Kong.April,2003:585-588.
[43]Williams P.S.and Alder M.D.Generic texture analysis applied tonewspaper segmentation.In Proceeding for ICNN96,Washington DC.June,1996.3:1664-1669.
[2]章毓晋.国际标准MPEG-7问答.中国图像图形学报.2000.5B(11-12):7-10.
[3]Faloutsos C et al.Efficient and Effective Querying by Image Content.Research Report RJ9453(B3704).1993.IBM Research Division.SanJose,Califor J Intell Information Systems.1994.3:231-262.
[4]Flickner M et al.Querying by Image and Video Content:The QBICSystem.Computer.1995:23-32.
[5]Niblack W et al.The QBIC Project:Querying Images by Content UsingColor,Texture and Shape.1993.SPIE,1908:173-187.
[6]庄越挺.智能多媒体信息分析与检索的研究[D].浙江大学博士论文.1998.
[7]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,(5):139-141.
[8]王绍棣.基于内容特征的图像和视频检索系统[J].计算机工程与应用.2001,(7):113-117.
[9]Greg Pass,Ramin Zabih,and Justin Miller.Comparing images usingcolor coherence vectors.Proc.ACM Conf.On Multimedia.1996.
[10]Huang J,Kumar S,Mitra M,Zhu W J,and Zabih R.Image indexingusing color correlogram.Proc.of IEEE Conf.On Computer Vision andPattern Recognition.1997.
[11]Smith J R and Chang S F.Local color and texture extraction and spatialquery.Proc.IEEE Int.Conf.on Image Proc.1996
[12]Ma W Y,Deng Yining,and Manjunath B.S..Tools for texture/colorbased search of images.SPIE Int.Conf.3106,Human Vision andElectronic Imaging.Feb 1997.
[13]Haralick Robert M,Shanmugam K.Texture Features for ImageClassification.IEEE Trans.on Systems.Man,Cybernetics.1973,SMC-3(6):610-621.
[14]Tamura Hideyuki,Mori Shunji,and Yamawaki Takashi.Texture featurescorresponding to visual perception.IEEE Trans.on Sys,Man,andCyb.1978.SMC-8(6).
[15]Ma W.Y.and Manjunath B.S..A comparison of wavelet transformfeatures for texture image annotation.Proc.IEEE Int.Conf.on Image.1994.
[16]Pentland A,Picard R.W.and Sclaroff S.Photobook:Content-basedmanipulation of image databases.International Journal of ComputerVision.1996.
[17]Chuang Gene C.H.and Kuo C.C.Jay.Wavelet descriptor of planar curves:Theory and applications.IEEE Trans.Image Proc..January,1996,5(1):56-70.
[18]Kass M,Witkin A,Terzopoulos D.Snake:Active Contour Model.IJCV,1988,1(4):321-331.
[19]Rickman Rick and Stonham John.Content-based image retrieval usingcolor tuple histograms.Proc.SPIE Storage and Retrieval for Image andVideo Databases.1996.
[20]Stone H.S.,Li C.S.,Image matching by means of intensity and texturematching in the Fourier domain.Proc.SPIE.1996,Vol.2670:337-349.
[21]Celentano A.and Lecce V.D..A FFT based technique for imagesignature generation.Proc.of SPIE:Storage and Retrieval for Image andVideo Databases V.Feb 1997,Vol.3022:457-466.
[22]Smith J.R.and S.F.Transform features for texture classification anddiscrimination in large image databases.Proc of IEEE Intl.Conf on ImageProcessing.1994.Vol.3:407-411.
[23]Reeves R.,Kubik K.and Osberger W..Texture characterization ofcompressed aerial images using DCT coefficients.Proc.of SPIE:Storageand Retrieval for Image and Video DatabasesV.Feb,1997.Vol.3022:398-407,
[24]Shneier M.and Motaleb M.A..Exploiting the JPEG compression schemefor image retrieval.IEEE Trans on Pattern Analysis and MachineIntelligence.Aug 1996.Vol.18,No.8:849-853.
[25]Chang T.and Kuo C.C.J..Texture analysis and classification withtree-structured wavelet transform.IEEE Trans.on Image Processing.Oct1993.Vol.2,No.4:429-441.
[26]Manjunath B.S.and Ma W.Y..Texture features for browsing and retrieval of image data.IEEE Trans.On Pattern Analysis and MachineIntelligence.Aug 1996.Vol.18,No.8:837-841.
[27]Muneesawang P,Guan L.Multiresolution histogram indexing andrelevance feed back learning for image retrieval.In:Proceedings of IEEEon Image Processing Vancouver,BC,Canada.2000:526.
[28]Manjunath B.S.and Ma W.Y..Image Indexing Using a TextureDictionary.Proc.of SPIE Conf.on Image Storage and Archiving System,volume 2606.
[29]Amos Tversky.Features of similarity.Psychological review.1977.84(4):327-352.
[30]Simone Santini and Ramesh Jain.Similarity Match.IEEE Trans.onPattern Analysis and Machine Intelligence.1996.
[31]Furht B et al.Video and Image Processing in Multimedia Systems.Boston:Kluwer Academic Publishers.1995:226-270.
[32]Perry B,et al.Content-based Access to Multimedia Information.KluwerAcademic Publishers.1999.
[33]崔锦泰,程正兴,白居宪.小波分析导论.西安交通大学出版社.1995.
[34]Aksoy S,Haralick R M.Feature normalization and likelihood-basedsimilarity measures for image retrieval.PRL,2001,22(5):563-582.
[35]Hemami S S.Image compression-a review.In:Image Databases-Searchand Retrieval of Digital Imagery.Castelli V,Bergman L D,eds.JohnWiley&Sons,Inc.2002.Ch.8:211-239.
[36]Jacquin A E.Fractal image coding:A review.Proc.IEEE.1993.81(10):1451-1465.
[37]曾文曲,文有为,孙炜.分形小波与图像压缩.沈阳:东北大学出版社.2002年第一版:第二章.
[38]Sloan A D.Retrieving database contents by image recognition:Newfractal power.Advanced Imaging.1994.9(5):26-30.
[39]Zhang A,Cheng B,Acharya R.An approach to query-by-texture inimage database systems.In Proceedings of the SPIE Conference onDigital Image Storage and Archiving Systems,Philadelphia,USA.Oct,1995:338-349.
[40]Zhang Aidong,Cheng Biao,Acharya Raj,et al..Comparison of wavelettransforms and fractal coding in texture based image retrieval.InProceedings of the SPIE Conference on Visual Data Exploration andAnalysisⅢ,San Jose.Jan,1996:116-125.
[41]Schouten B,Zeeuw P.Image databases,scale and fractal transforms.Proceeding of ICIP,Canada.Sept 2000,vol.2:534-537.
[42]Pi M,Mandal M,Basu A.Image retrieval based on histogram of newfractal parameters.Proceeding of ICASSP,Hong Kong.April,2003:585-588.
[43]Williams P.S.and Alder M.D.Generic texture analysis applied tonewspaper segmentation.In Proceeding for ICNN96,Washington DC.June,1996.3:1664-1669.