(1.東南大學,影像科學與技術實驗室,江蘇省，南京市210096
　　                    2.東南大學,影像科學與技術實驗室,江蘇省，南京市210096
　　                   3.東南大學,影像科學與技術實驗室,江蘇省，南京市210096)
　　
　　
摘要：近年來，由于醫學成像技術的飛速發展，產生大量的數字化圖像。醫學圖像是醫生診斷的重要參考依據，因此如何對醫學圖像有效的標注和檢索變得越來越重要。本文提出利用圖學習的方法來標注醫學圖像。首先根據醫學圖像的特性進行預處理，然后將圖像信息和標注詞建立聯系，最后將標注詞從已標注圖像傳遞到未標注圖像。實驗結果證明，基于圖學習的醫學圖像標注方法明顯的提高了標注性能，是一種有效的方法。
關鍵詞：圖像標注；檢索；圖學習
中圖分類號：        文獻標識碼：        文章編號：
Medical Image Annotation Based On Graph Learning
　　　WANG Jie, SHU Hua-zhong, WANG Bin
　�。↖maging Science and Technology Laboratory，Southeast University, 210096, China .
　　WANG Jie, E-mail:wangjie28183@163.com;
　　SHU Hua-zhong, E-mail: shu.list@seu.edu.cn;
　　WANG Bin, E-mail: binwang@seu.edu.cn）
Abstract：In recent years, it produces large amounts of digitized images because of the rapid development of medical imaging technology. Medical images are important reference for doctors’ diagnosis, so how to annotate and retrieve medical images effectively is becoming increasingly important. In this paper, we come up with graph learning method to annotate medical images. First, do preprocessing according to the characteristics of medical images. Then contact images’ information with annotation words. Finally pass the marked word from the labeled images to unlabeled images. The experimental results show that it is an effective method of graph learning method to annotate medical images, and the performance is improved.
Key words：Image annotation；Retrieval；Graph learning

參考文獻 (References)
 [1]H Lilian Tang，Rudolf Hanka，Horace H S. Histological Image Retrieval Based on Semantic       Content Analysis[J].IEEE Transactions on information technology in biomedicine，2003，3，7(1)
[2] Mori Y, Takahashi H, Oka R. Image-to-word transformation based on dividing and vector quantizing images with words[C].Orlando,Oct.1999
[3] Oliva A,Torralba A.Modeling the shape of the scene:A holistic representation of the spatial envelope[J].Int. J.Comput. Vision,2001,42(3):145-175.
[4] Oliva A,Torralba A B.Scene-centered description from spatial envelope properties[C].2002:263-272.
[5] Tang J,Lewis P H.A study of quality issues for image auto-annotation with the Corel dataset[J].IEEE Trans.on Circuits and Systems for Video Technology,2007,17(3):384-389
[6] Cusano C,Ciocca G,Schettini R.Image annotation using SVM [C]. USA,Feb.2004:330-338
[7] Gustavo C,Nuno V.A database centric view of semantic image annotation and retrieval[C]. Salvador,Brazil,Aug.2005:559-566
[8] Carneiro G,Chan A B,Moreno P J,et al.Supervised learning of semantic classes for image annotation and retrieval[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence,2007,29 (3):394-410
[9] Jeon J,Lavrenko V,Manmatha R.Automatic image annotation and retrieval using cross-media relevance models[C]. Toronto, Canada, July 2003:119-126
[10] Lavrenko V,Manmatha R,Jeon J.A model for learning the semantics of pictures[C]∥Proc.of Advances in Neural Information Processing Systems (NIPS’03).2003
[11] Feng S,Manmatha R,Lavrenko V.Multiple bernoulli relevance models for image and video annotation[C]. Washington DC,USA,June 2004:1002-1009
[12] Pan J Y,Yang H J,Pinar D.Automatic multimedia cross-modal correlation discovery[C]. August 2004:653-658
[13] CHANG S G, YU B, VETTERLIM. Adaptivewavelet thresholding for image denoising and compression [J]. IEEE Transactions on Image Processing, 2000, 9(9): 1532-1546.
[14] JOBSON D J, RAHMAN Z U, WOODELL G A. A multiscale retinex forbridging the gap between color images and the human observation of scenes [J]. IEEE Transactions on Image Processing,1997, 6(7): 965-976.
[15] WANG HAI-TAO, LI S Z, WANG YANG-SHENG. Face recognition under varying lighting conditions using self quotient image[C]. IEEE ComputerSociety, 2004: 819-824.

作者簡介：
王潔，女，漢族，就讀于東南大學，碩士研究生，主要研究方向為圖像檢索和圖像標注。