扫频光学相干层析角膜图像轮廓自动提取算法

汪毅; 刘珊珊; 张玮茜; 蔡怀宇; 陈晓冬

doi:10.7498/aps.68.20190731

摘要

在扫频光学相干层析系统中, 远心扫描模式造成角膜图像中存在伪影、部分结构缺失及低信噪比区域, 影响了角膜轮廓提取的精度. 针对该问题, 本文提出了一种针对低质量角膜图像的轮廓自动提取算法. 该算法首先依据图像标准差分布将图像划分为高、低信噪比区域; 针对高信噪比区域, 通过峰值点定位法获取角膜轮廓; 针对低信噪比区域, 通过连续帧图像间配准叠加实现图像增强, 为低信噪比区域提供参考轮廓点, 再通过权衡参考轮廓点与局部直线拟合结果的优劣, 实现角膜轮廓定位; 最后, 通过全局多项式拟合实现对全区域的角膜整体轮廓信息. 对光学眼模型进行实验, 结果表明, 与已有算法相比, 本文算法对角膜轮廓的提取精度平均提高了4.9%.

关键词:

Abstract

In a swept source-optical coherence tomography system, the telecentric scanning mode gives rise to central saturation artifacts,partial structural loss, and low SNR (signal-to-noise ratio) area in the corneal image, which affects the accuracy of corneal contour extraction. In order to solve this problem, in this paper we propose an automatic extraction algorithm for corneal image of low quality. This algorithm divides the image into high and low SNR region according to the standard deviation distribution of the cornea image. For the high SNR region, we localize the peak point to extract the contour. For the low SNR region, image enhancement is achieved by the registration and superposition of successive frames, which provides reference contour points for low SNR areas. Then corneal contour localization is achieved by weighing the advantages and disadvantages of reference contour points and local line fitting results. Finally, global polynomial fitting is used to achieve the whole corneal contour information. Experiments on the optical eye model show that comparing with the existing algorithms, the accuracy of corneal contour extraction is improved by 4.9% on average.

Keywords:

optical coherence tomography /
swept-source optical coherence tomography /
cornea /
contour extraction

作者及机构信息

天津大学精密仪器与光电子工程学院, 光电信息技术教育部重点实验室, 天津　300072

通信作者: 陈晓冬, xdchen@tju.edu.cn

基金项目: 国家重点研发计划(批准号: 2017YFC0109901)和天津市自然科学基金(批准号: 15JCQNJC14200)资助的课题

Authors and contacts

Key Laboratory of Opto-Electronics Information Technology of Ministry of Education, College of Precision Instrument and Opto-Electronic Engineering, Tianjin University, Tianjin 300072, China

Corresponding author: Chen Xiao-Dong, xdchen@tju.edu.cn

Funds: Project supported by the National Key Research and Development Projectof China (Grant No. 2017YFC0109901) and the Natural Science Foundation of Tianjin, China (Grant No. 15JCQNJC14200)

文章全文

参考文献

[1]	Wang J F, Zheng W, Lin K, Huang Z W 2016 Opt. Lett. 41 3045 Google Scholar
[2]	Kumar A, Baumann B, Hafner J, Ginner L, Augustin M, Salas M, Pircher M, Leitgeb R, Prager S, Schmidterfurth U 2016 Biomed. Opt. Express 8 207
[3]	Grulkowski I, Liu J J, Potsaid B, Jayaraman V, Jiang J, Fujimoto J G, Cable A E 2013 Opt. Lett. 38 673 Google Scholar
[4]	Polans J, Cunefare D, Cole E, Keller B, Mettu P S, Cousins S W, Allingham M J, Izatt J A, Farsiu S 2017 Opt. Lett. 42 17 Google Scholar
[5]	Camino A, Jia Y L, Liu G J, Wang J, Huang D 2017 Biomed. Opt. Express 8 3053 Google Scholar
[6]	Kaluzny B J, Karnowski K, Szkulmowski M, Wojtkowski M, Gora M 2011 Biomed. Opt. Express 2 2709 Google Scholar
[7]	Lawman S, Dong Y, Williams B M, Romano V, Kaye S, Harding S P, Willoughby C, Shen Y C, Zheng Y L 2016 Opt. Express 24 12395 Google Scholar
[8]	Lawman S, Madden P W, Vito R, Romano V, Dong Y, Mason S, Williams B M, Kaye S B, Willoughby C E, Harding S P, Shen Y C 2017 Biomed. Opt. Express 8 5579 Google Scholar
[9]	Wang L L, Xiong Q Z, Ge X, Bo E, Xie J, Liu X Y, Yu X J, Wang X H, Wang N S, Chen S, Wu X, Liu L B 2019 Opt. Express 27 1298 Google Scholar
[10]	Williams D, Zheng Y L, Bao F J, Elsheikh A 2013 J. Biomed. Opt. 18 056003
[11]	Li Y, Shekhar R, Huang D 2002 Proc. SPIE 4684 167 Google Scholar
[12]	Li Y, Shekhar R, Huang D 2006 Ophthalmology 113 792 Google Scholar
[13]	Shu P, Sun Y 2012 J. Innov. Opt. Heal. Sci. 5 9
[14]	LaRocca F, Chiu S J, McNabb R P, Kuo A N, Izatt J A, Farsiu S 2011 Biomed. Opt. Express 2 1524 Google Scholar
[15]	Keller B, Draelos M, Tang G, Farsiu S, Kuo A N, Hauser K, Izatt J A 2018 Biomed. Opt. Express 9 2716 Google Scholar
[16]	Santos V A, Schmetterer L, Stegmann H, Pfister M, Messner A, Schmidinger G, Garhofer G, Werkmeister R M 2019 Biomed. Opt. Express 10 622 Google Scholar
[17]	Otsu N 2007 IEEE. Trans. Syst. Man. Cybern. 9 62
[18]	Doughty M J, Zaman M L 2000 Surv. Ophthalmol. 44 367 Google Scholar
[19]	Gifford P, Ahmed A, Swarbrick H A 2011 Invest. Ophth. Vis. Sci. 52 3648 Google Scholar
[20]	Atchison D A, Jones C E, Schmid K L, Nicola P, Pope J M, Strugnell W E, Riley R A 2004 Invest. Ophth. Vis. Sci. 45 3380 Google Scholar
[21]	Wang Q, Liu W W, Wu Y L, Ma Y, Zhao G Q 2017 Clin. Exp. Optom. 100 250
[22]	Saenzfrances F, Gonzalezpastor E, Borregosanz L, Jerezfidalgo M, Martinezdelacasa J, Mendezhernandez C, Santosbueso E, Fernandezvidal A, Garciasanchez J, Garciafeijoo J 2012 J. FR. Ophtalmol. 35 333 Google Scholar
[23]	Read S A, Collins M J 2009 Optometry & Vision Science Official Publication of the American Academy of Optometry 86 170
[24]	Zaki F, Wang Y H, Su H, Yuan X, Liu X 2017 Biomed. Opt. Express 8 2720 Google Scholar
[25]	张强, 那彦, 李建军 2006 应用光学 27 4 Zhang Q, Na Y, Li J J 2006 J. Appl. Opt. 27 4

施引文献

图 1 角膜轮廓提取算法流程图

Fig. 1. Flow chart of corneal contour extraction algorithm

下载: 全尺寸图片幻灯片

图 2 角膜图像高、低信噪比区域划分　(a) 列间标准差平滑结果; (b) 角膜图像高、低信噪比区域划分结果

Fig. 2. Division between high and low SNR (signal-to-noise ratio) regions of corneal image: (a) Smoothing result of standard deviation between columns; (b) division results of high and low SNR regions of corneal image.

下载: 全尺寸图片幻灯片

图 3 高信噪比区域的轮廓提取结果　(a) 轮廓点初步提取结果; (b) 轮廓点精确提取结果

Fig. 3. Contour extraction results of high SNR region: (a) Preliminary extraction result of contour points; (b) accurate extraction result of contour points.

下载: 全尺寸图片幻灯片

图 4 角膜整体轮廓提取过程　(a) OTSU算法处理结果; (b) 中值滤波处理结果

Fig. 4. Extraction process of the overall cornea contour: (a) Processing result by OTSU algorithm; (b) processing result by median filtering.

下载: 全尺寸图片幻灯片

图 5 低信噪比区域角膜轮廓建模及提取结果　(a)低信噪比区域表面轮廓点建模结果; (b)低信噪比区域轮廓提取结果

Fig. 5. Modeling and extraction results of corneal contour in low SNR region: (a) Modeling results of surface contour points in low SNR region; (b) contour extraction result of low SNR region.

下载: 全尺寸图片幻灯片

图 6 角膜完整轮廓的提取过程　(a)角膜上下表面轮廓点提取结果; (b) 角膜轮廓拟合结果

Fig. 6. Extraction process of the complete cornea contour: (a) Extraction results of the contour points in the upper and lower cornea surfaces; (b) fitting result of the cornea contour

下载: 全尺寸图片幻灯片

图 7 两种算法效果对比　(a)两种算法轮廓提取结果; (b) 两种算法角膜厚度平均计算结果

Fig. 7. Comparison of the effects of the two algorithms: (a) Results of the contour extraction of the two algorithms; (b) results of the corneal thickness calculated by the two algorithms.

下载: 全尺寸图片幻灯片

表 1 两种算法对角膜轮廓平均提取精度对比

Table 1. Comparison of the accuracy of two algorithmsfor contour extraction of high and low SNR regions.

评价指标文献[14] 本文算法精度提高值/%

平均误差值/pixels 2.8 0.9 4.9
偏差率/% 7.2 2.3

下载: 导出CSV

[1]	Wang J F, Zheng W, Lin K, Huang Z W 2016 Opt. Lett. 41 3045 Google Scholar
[2]	Kumar A, Baumann B, Hafner J, Ginner L, Augustin M, Salas M, Pircher M, Leitgeb R, Prager S, Schmidterfurth U 2016 Biomed. Opt. Express 8 207
[3]	Grulkowski I, Liu J J, Potsaid B, Jayaraman V, Jiang J, Fujimoto J G, Cable A E 2013 Opt. Lett. 38 673 Google Scholar
[4]	Polans J, Cunefare D, Cole E, Keller B, Mettu P S, Cousins S W, Allingham M J, Izatt J A, Farsiu S 2017 Opt. Lett. 42 17 Google Scholar
[5]	Camino A, Jia Y L, Liu G J, Wang J, Huang D 2017 Biomed. Opt. Express 8 3053 Google Scholar
[6]	Kaluzny B J, Karnowski K, Szkulmowski M, Wojtkowski M, Gora M 2011 Biomed. Opt. Express 2 2709 Google Scholar
[7]	Lawman S, Dong Y, Williams B M, Romano V, Kaye S, Harding S P, Willoughby C, Shen Y C, Zheng Y L 2016 Opt. Express 24 12395 Google Scholar
[8]	Lawman S, Madden P W, Vito R, Romano V, Dong Y, Mason S, Williams B M, Kaye S B, Willoughby C E, Harding S P, Shen Y C 2017 Biomed. Opt. Express 8 5579 Google Scholar
[9]	Wang L L, Xiong Q Z, Ge X, Bo E, Xie J, Liu X Y, Yu X J, Wang X H, Wang N S, Chen S, Wu X, Liu L B 2019 Opt. Express 27 1298 Google Scholar
[10]	Williams D, Zheng Y L, Bao F J, Elsheikh A 2013 J. Biomed. Opt. 18 056003
[11]	Li Y, Shekhar R, Huang D 2002 Proc. SPIE 4684 167 Google Scholar
[12]	Li Y, Shekhar R, Huang D 2006 Ophthalmology 113 792 Google Scholar
[13]	Shu P, Sun Y 2012 J. Innov. Opt. Heal. Sci. 5 9
[14]	LaRocca F, Chiu S J, McNabb R P, Kuo A N, Izatt J A, Farsiu S 2011 Biomed. Opt. Express 2 1524 Google Scholar
[15]	Keller B, Draelos M, Tang G, Farsiu S, Kuo A N, Hauser K, Izatt J A 2018 Biomed. Opt. Express 9 2716 Google Scholar
[16]	Santos V A, Schmetterer L, Stegmann H, Pfister M, Messner A, Schmidinger G, Garhofer G, Werkmeister R M 2019 Biomed. Opt. Express 10 622 Google Scholar
[17]	Otsu N 2007 IEEE. Trans. Syst. Man. Cybern. 9 62
[18]	Doughty M J, Zaman M L 2000 Surv. Ophthalmol. 44 367 Google Scholar
[19]	Gifford P, Ahmed A, Swarbrick H A 2011 Invest. Ophth. Vis. Sci. 52 3648 Google Scholar
[20]	Atchison D A, Jones C E, Schmid K L, Nicola P, Pope J M, Strugnell W E, Riley R A 2004 Invest. Ophth. Vis. Sci. 45 3380 Google Scholar
[21]	Wang Q, Liu W W, Wu Y L, Ma Y, Zhao G Q 2017 Clin. Exp. Optom. 100 250
[22]	Saenzfrances F, Gonzalezpastor E, Borregosanz L, Jerezfidalgo M, Martinezdelacasa J, Mendezhernandez C, Santosbueso E, Fernandezvidal A, Garciasanchez J, Garciafeijoo J 2012 J. FR. Ophtalmol. 35 333 Google Scholar
[23]	Read S A, Collins M J 2009 Optometry & Vision Science Official Publication of the American Academy of Optometry 86 170
[24]	Zaki F, Wang Y H, Su H, Yuan X, Liu X 2017 Biomed. Opt. Express 8 2720 Google Scholar
[25]	张强, 那彦, 李建军 2006 应用光学 27 4 Zhang Q, Na Y, Li J J 2006 J. Appl. Opt. 27 4

[1]	钱黄河, 王迪, 韩涛, 丁志华. 基于复数主从光学相干层析成像相位信息的离散界面快速定位方法. 物理学报, 2022, 71(21): 214202. doi: 10.7498/aps.71.20220444
[2]	吴彤, 霍文麒, 黄蕴智, 王吉明, 顾晓蓉, 路元刚, 赫崇君, 刘友文. 用于内窥光学相干层析成像的小型化预标定Lissajous扫描光纤探头. 物理学报, 2021, 70(15): 150701. doi: 10.7498/aps.70.20210151
[3]	李海, 邹健, 邵彬, 陈雨, 华臻. 库的量子关联相干辅助系统能量提取的研究. 物理学报, 2019, 68(4): 040201. doi: 10.7498/aps.68.20181525
[4]	胡喆皓, 上官紫微, 邱建榕, 杨珊珊, 鲍文, 沈毅, 李鹏, 丁志华. 基于受激辐射信号的谱域光学相干层析分子成像方法. 物理学报, 2018, 67(17): 174201. doi: 10.7498/aps.67.20171738
[5]	吴彤, 孙帅帅, 王绪晖, 王吉明, 赫崇君, 顾晓蓉, 刘友文. 基于最优化线性波数光谱仪的谱域光学相干层析成像系统. 物理学报, 2018, 67(10): 104208. doi: 10.7498/aps.67.20172606
[6]	王毅, 郭哲, 朱立达, 周红仙, 马振鹤. 基于谱域相位分辨光学相干层析的纳米级表面形貌成像. 物理学报, 2017, 66(15): 154202. doi: 10.7498/aps.66.154202
[7]	樊金宇, 高峰, 孔文, 黎海文, 史国华. 多面转镜激光器扫频光学相干层析成像系统的全光谱重采样方法. 物理学报, 2017, 66(11): 114204. doi: 10.7498/aps.66.114204
[8]	马振鹤, 窦世丹, 马毓姝, 刘健, 赵玉倩, 刘江红, 吕江涛, 王毅. 基于光学相干层析成像的早期鸡胚心脏径向应变测量. 物理学报, 2016, 65(23): 235202. doi: 10.7498/aps.65.235202
[9]	上官紫微, 沈毅, 李鹏, 丁志华. 扫频光学相干层析成像系统的波数校正与相位测量研究. 物理学报, 2016, 65(3): 034201. doi: 10.7498/aps.65.034201
[10]	潘聪, 郭立, 沈毅, 严雪过, 丁志华, 李鹏. 基于界面信号的扫频光学相干层析成像系统相位矫正方法. 物理学报, 2016, 65(1): 014201. doi: 10.7498/aps.65.014201
[11]	唐弢, 赵晨, 陈志彦, 李鹏, 丁志华. 超高分辨光学相干层析成像技术与材料检测应用. 物理学报, 2015, 64(17): 174201. doi: 10.7498/aps.64.174201
[12]	赵晨, 陈志彦, 丁志华, 李鹏, 沈毅, 倪秧. 线照明并行谱域光学相干层析成像系统与缺陷检测应用研究. 物理学报, 2014, 63(19): 194201. doi: 10.7498/aps.63.194201
[13]	张品, 梁艳梅, 常胜江, 范海伦. 基于能量最小化的肾脏计算断层扫描图像分割方法. 物理学报, 2013, 62(20): 208701. doi: 10.7498/aps.62.208701
[14]	刘国忠, 周哲海, 邱钧, 王晓飞, 刘桂礼, 王瑞康. 幅值和相位配准技术及其在光学相干层析血流成像中的应用. 物理学报, 2013, 62(15): 158702. doi: 10.7498/aps.62.158702
[15]	鲍文, 丁志华, 王川, 梅胜涛. 基于相位敏感谱域光学相干层析术的潜指纹获取方法. 物理学报, 2013, 62(11): 114202. doi: 10.7498/aps.62.114202
[16]	王凯, 曾焱, 丁志华, 孟婕, 史国华, 张雨东. 谱域光学相干层析系统中基于解卷积方法的像质优化. 物理学报, 2010, 59(4): 2471-2478. doi: 10.7498/aps.59.2471
[17]	杨亚良, 丁志华, 王凯, 吴凌, 吴兰. 全场光学相干层析成像系统的研制. 物理学报, 2009, 58(3): 1773-1778. doi: 10.7498/aps.58.1773
[18]	张耘. 极化子荧光及其断层扫描对Ti:LiNbO3光波导表征研究. 物理学报, 2007, 56(1): 280-284. doi: 10.7498/aps.56.280
[19]	梁艳梅, 周大川, 孟凡勇, 王明伟. 一种新型的专用于光学相干层析系统的宽带光纤光源. 物理学报, 2007, 56(6): 3246-3250. doi: 10.7498/aps.56.3246
[20]	贾亚青, 梁艳梅, 朱晓农. 光学相干层析信号的模拟分析与计算. 物理学报, 2007, 56(7): 3861-3866. doi: 10.7498/aps.56.3861

计量

文章访问数: 8268
PDF下载量: 68
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板

扫频光学相干层析角膜图像轮廓自动提取算法