Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Acta Electronica Sinica》 2017-08
Add to Favorite Get Latest Update

A Mean Shift Algorithm and Interpolation Image Restoration Algorithm Based Method for Metal Artifact Reduction

LI Jian-feng;LIAO Sheng-hui;MEI Chu-xuan;College of Information Science and Enginnering,Jishou University;College of Information Science and Enginnering,Central South University;  
In recent years,CT imaging technique has been widely used in clinical medicine domain. However for the patient with metallic implants,there is most likely bright and dark artifacts around metal objects due to beam hardening and other reasons,which will reduce the quality of the image and affect the accuracy of physician diagnosis. In order to reduce the metal artifacts in CT images,we address a method based on Mean Shift algorithm and interpolation image restoration algorithm in this paper,and the basic flowis as follows. Firstly,adaptive Mean Shift algorithm was used to smooth noise and the mild artifacts for preprocessing. Secondly,using Mean Shift simplified algorithm metal objects were segmented quickly and accurately. Thirdly,the prior image was generated by interpolated image with anatomical structures restoration,then the original projection data was replaced with the prior projection data to obtain the final corrected image. After the contrast experiment,the method in this paper not only removes metal artifacts remarkably but also protects the original anatomical structures,showing that our method achieves better appearance.
【Key Words】: metal artifact reduction Mean Shift algorithm interpolation prior image projection completion
【Fund】: 国家自然科学基金(No.61562029 No.61363073 No.61262032);; 湖南省自然科学基金(No.2015JJ3100);; 吉首大学学成返校取得博士学位、学历人员科研资助经费研究项目(吉大人通[2014]39号)
【CateGory Index】: R814.42;TP391.41
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved