杨帆,何艳,邓振生,严昂
YANG Fan,HE Yan,DENG Zhensheng,YAN Ang

• 1:贵州医科大学生物医学工程教研室
• 2:中南大学生物医学工程研究所
• 3:中南大学湘雅二医院设备科

摘要(Abstract)：

目的:探讨一种基于图像处理技术的全脊柱及下肢X线图像自动拼接算法的可行性。方法:采用医学图像处理技术,通过相位相关算法确定粗略重叠区域,利用Harris算法提取特征点和角点邻域灰度信息实现图像的精确配准,对40组多次拍摄的脊柱或下肢X线图像进行算法拼接,并与设备自带软件拼接的图像进行比较。结果:对40组多次拍摄的脊柱或下肢X线图像进行算法拼接,误差4 mm内拼接成功率达到95%,高于自带软件拼接结果。结论:算法拼接法能有效实现多次拍摄的脊柱或下肢X线图像拼接。
Objective: To explore the feasibility of automatic stitching method for full-spine and lower limbs X-ray images based on image processing technique. Methods: Medical image processing technology was adopted in this research. Phase correlation algorithm was used to identify rough areas,and Harris algorithm to extract features and corner neighborhood gray scale information to actualize accurate image registration. 40 pairs of spine or lower limbs X-ray images were used to conduct algorithm stitching and compared with the result of stitching software. Results: 40 pairs of spine or lower limbs X-ray images were used to conduct algorithm stitching,and the stitching success rate with deviation within4 mm was 95%,which was higher than that of stitching software. Conclusion: The algorithm stitching method can effectively actualize image stitching of spine or lower limb X-ray images.

关键词(KeyWords)： X线图像;相位相关;角点;自动拼接
X-ray images;phase correlation;corner;auto-stitching

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金地区科学基金项目(81460206)

作者(Author): 杨帆,何艳,邓振生,严昂
YANG Fan,HE Yan,DENG Zhensheng,YAN Ang

DOI: 10.19367/j.cnki.1000-2707.2016.04.008

参考文献(References)：

• [1]白亚妮,贺洪德,邓振生.CR在双下肢全长投照技术中的应用[J].实用放射学杂志,2006(9):1141-1142.
• [2]Dewaele P,Vuylsteke P,Schoeters EP,et al.Full-leg/full-spine image stitching:a new and accurate CR-based imaging technique[C].Medical Imaging 1999:131-138.
• [3]Yaniv Z,Joskowicz L.Long bone panoramas from fluoroscopic x-ray images[J].IEEE transactions on medical imaging,2004(1):26-35.
• [4]Gooben A,Schluter M,Pralow T,et al.A stitching algorithm for automatic registration of digital radiographs[C].In A.Campilho and M.Kamel,editors,Image Analysis and Recognition,Lecture Notes in Computer Science,2008:854-862.
• [5]Markus G,Wilfried B,Berned L,et al.An Algorithm for Automatic Stitching of CR X-ray Images[J].Advances in Medical Engineering,2007(114):193-198.
• [6]Wang L,Traub J,Weidert S,et al.Parallax-Free Long Bone X-ray Image Stitching[C].Medical Image Computing and Computer-Assisted Intervention(MICCAI),London,UK,2009:173-180.
• [7]徐亮,覃涛.DR图像手动拼接技术在全膝关节置换术中的应用[J].华中科技大学学报:医学版,2014(4):456-458.
• [8]李振宇,苏文姬.DR图像拼接技术的临床应用[J].医学理论与实践,2012(21):2678-2679.
• [9]张旭峰,严壮志,Wanigasekar NR,等.基于拼接技术的医学X光片全景成像[J].航天医学与医学工程,2005(5):360-364.
• [10]王双玲.基于双向异步投票策略角点匹配的X线医学图像拼接[D].长沙:中南大学,2011.
• [11]赵宏,赵凯,江春花,等.旋转不变的X线图像相关度拼接[J].中国图象图形学报,2013(6):654-660.
• [12]Kuglin CD,Hines DC.The phase correlation image alignment method[C].IEEE International Conference on Cybemetics and Society,1975:163-165.
• [13]Foroosh H,Zerubia JB,Berthod M.Extension of Phase Correlation to Subpixel Registration[J].IEEE Transactions on Image Processing,2002(3):188-200.
• [14]Harris C,Stephens M.A Combined Corner and Edge Detector[C].Proc.of 4th Alvey Vision Conference,1988:147-152.
• [15]Noble J A.Finding corners[J].Image and vision Computin,1988(2):121-128.
• [16]Fischler MA,Bolles RC.Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J].Communication of ACM,1981(6):381-395.
• [17]Szeliski R.Video Mosaics for Virtural Environments[J].IEEE Computer Graphics and Applications,1996(2):22-29.

文章评论(Comment)：