| 题名 | 高像质频域光学相干层析成像关键技术研究 |
| 作者 | 张行 |
| 学位类别 | 硕士 |
| 答辩日期 | 2015 |
| 授予单位 | 中国科学院上海光学精密机械研究所 |
| 导师 | 李中梁 |
| 关键词 | 光学相干层析成像 散斑抑制 色散补偿 成像质量 |
| 其他题名 | Research on Key Technology of High Image Quality for Fourier Domain Optical Coherence Tomography |
| 中文摘要 | 光学相干层析成像（Optical Coherence Tomography，OCT）是一种可对生物组织进行非侵入、高分辨率、实时成像的生物医学光学成像技术，已广泛地应用于眼科、心血管、皮肤、消化道和呼吸道等临床诊断领域。OCT系统的成像质量直接决定了测量结果能否更真实可靠地反映所测生物组织的内部结构信息。与时域OCT相比，频域OCT在成像速度和灵敏度方面有明显的优势，更适合于对生物组织进行成像。然而，频域OCT存在散斑噪声和色散等问题，严重影响了系统的成像质量：散斑噪声会降低图像对比度，模糊组织结构的特征细节；色散会导致系统点扩散函数的展宽，降低系统轴向分辨率。本论文针对影响频域OCT系统成像质量的散斑噪声和色散问题，提出了两种改进技术。主要工作包括以下两个方面： 1. 提出一种基于两步配准策略的多幅图像配准平均散斑噪声抑制技术。图像的配准精度会影响散斑噪声抑制的效果，该技术将全局配准与局部配准相结合，可有效提高配准精度。首先通过全局配准补偿图像间的整体移动，全局配准基于刚体变换模型能够补偿图像间的平移和旋转。然后进行局部配准，对图像的每一个A-scan通过计算互相关值采用基于图论的算法进行对准，并且进一步针对图像离散化造成的断层现象，采用像素细分的方法提高图像平滑度。该技术不需要利用任何视网膜结构的边界信息来辅助配准。对人眼视网膜的成像实验表明，利用该技术能够有效地抑制散斑噪声，提高图像信噪比和对比度噪声比。 2. 提出一种基于多窗口迭代空间滤波的深度分辨色散补偿方法。采用窗口迭代空间滤波的方法由粗到细的补偿系统中色散，并且对样品不同深度处采用不同的色散系数进行补偿，达到深度分辨色散补偿的效果。该方法避免了采用单一阈值确定的窗口宽度不合适对深度分辨色散补偿效果的影响，可以获得更佳的色散补偿效果。仿真实验表明，采用该方法可以精确的补偿样品不同深度处的色散，提高频域OCT系统的轴向分辨率。 |
| 英文摘要 | Optical coherence tomography (OCT) is a high-resolution, noninvasive, real-time, 3D imaging technology with great potential in both clinical and fundamental research applications in many areas, especially in the area of ophthalmology. Fourier domain OCT (FDOCT) has been widely attracted in the biomedical imaging field due to its higher sensitivity and imaging speed compared to conventional time domain OCT (TDOCT). However, FDOCT suffers from the problems of speckle noise (i.e., signal-degrading speckle) and dispersion, which can severely influence the image quality. Speckle noise reduces the contrast of obtained images and blurs the feature details. Dispersion expands the point spread function of the system and reduces the axial resolution. In view of the above problems, two techniques are presented in this dissertation to enhance the image quality. The main researches of this dissertation are as follows. 1. A two-step image registration method that combines global and local registrations is proposed for speckle reduction by the averaging of multiple B-scans. The method begins with a global registration to compensate for overall motion, which is estimated based on the rigid transformation model involving translation and rotation. Then each A-scan is aligned by cross-correlation using a graph-based algorithm, which followed by a pixel subdivision method to improve smoothness in local registration. The method does not rely on any information about the retinal layer boundaries. We have applied this method to the registration of macular optical coherence tomography images. The results show the reduction of speckle noise and the enhanced visualization of layer structures. A signal-to-noise ratio improvement of nearly the square root of the number of averaged B-scans and a contrast-to-noise ratio improvement of around 11 are achieved through our method. 2. A depth-resolved dispersion compensation method based on multi-window iteration is proposed. The proposed method can compensate the dispersion from coarse to fine in different imaging depth. The use of multi-window can solve the problem that the window width determined by a single threshold may be improper for a depth-resolved dispersion compensation. The results of numerical simulation show that with the proposed method the dispersion is compensated effectively and the axial resolution is improved. |
| 语种 | 中文 |
| 内容类型 | 学位论文 |
| 源URL | [http://ir.siom.ac.cn/handle/181231/16919]  |
| 专题 | 上海光学精密机械研究所_学位论文 |
| 推荐引用方式 GB/T 7714 | 张行. 高像质频域光学相干层析成像关键技术研究[D]. 中国科学院上海光学精密机械研究所. 2015. |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论