• 主题
高级搜索  >
历史搜索 清空历史
首页> 美国卫生研究院文献>Biomedical Optics Express >Three-dimensional curvelet-based dictionary learning forspeckle noise removal of optical coherence tomography
【2h】

Three-dimensional curvelet-based dictionary learning forspeckle noise removal of optical coherence tomography

机译：基于三维曲线波的字典学习光学相干断层扫描的斑点噪声去除

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务，本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文，但由于OA文献来源多样且变更频繁，仍可能出现获取不到、文献不完整或与标题不符等情况，如果获取不到我们将提供退款服务。请知悉。
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Optical coherence tomography (OCT) is a recently emerging non-invasivediagnostic tool useful in several medical applications such asophthalmology, cardiology, gastroenterology and dermatology. One ofthe major problems with OCT pertains to its low contrast due to thepresence of multiplicative speckle noise, which limits thesignal-to-noise ratio (SNR) and obscures low-intensity and smallfeatures. In this paper, we recommend a new method using the 3Dcurvelet based K-times singular value decomposition (K-SVD) algorithmfor speckle noise reduction and contrast enhancement of theintra-retinal layers of 3D Spectral-Domain OCT (3D-SDOCT) images. Inorder to benefit from the near-optimum properties of curvelettransform (such as good directional selectivity) on top of dictionarylearning, we propose a new plan in dictionary learning by using thecurvelet atoms as the initial dictionary. For this reason, thecurvelet transform of the noisy image is taken and then the noisycoefficients matrix in each scale, rotation and spatial coordinates ispassed through the K-SVD denoising algorithm with predefined 3Dinitial dictionary that is adaptively selected from thresholdedcoefficients in the same subband of the image. During the denoising ofcurvelet coefficients, we can also modify them for the purpose ofcontrast enhancement of intra-retinal layers. We demonstrate theability of our proposed algorithm in the speckle noise reduction of 17publicly available 3D OCT data sets, each of which contains 100B-scans of size 512×1000 with and without neovascularage-related macular degeneration (AMD) images acquired using SDOCT,Bioptigen imaging systems. Experimental results show that animprovement from 1.27 to 7.81 in contrast to noise ratio (CNR), andfrom 38.09 to 1983.07 in equivalent number of looks (ENL) is achieved,which would outperform existing state-of-the-art OCT despecklingmethods.
机译：光学相干断层扫描（OCT）是一种新兴的非侵入性技术诊断工具可用于多种医疗应用，例如眼科，心脏病学，肠胃病学和皮肤病学。之一OCT的主要问题是由于斑点斑点噪声的存在，限制了信噪比（SNR）并掩盖了低强度和小尺寸特征。在本文中，我们建议使用3D的新方法基于Curvelet的K时间奇异值分解（K-SVD）算法用于减少斑点噪声和增强对比度3D光谱域OCT（3D-SDOCT）图像的视网膜内层。在为了从Curvelet的近乎最佳特性中受益在字典顶部进行变换（例如良好的方向选择性）学习，我们通过使用Curvelet原子作为初始字典。因此，进行噪声图像的Curvelet变换，然后进行噪声每个尺度，旋转和空间坐标系中的系数矩阵为通过具有预定义3D的K-SVD去噪算法从阈值中自适应选择的初始字典图像相同子带中的系数。在去噪期间Curvelet系数，我们也可以出于以下目的对其进行修改增强视网膜内层的对比度。我们展示了我们提出的算法在减少斑点噪声方面的能力为17公开可用的3D OCT数据集，每个数据集包含100个带有和不带有新生血管的512×1000大小的B扫描使用SDOCT获得的年龄相关性黄斑变性（AMD）图像，Bioptigen成像系统。实验结果表明与噪声比（CNR）相比，从1.27提高到7.81，并且从38.09到1983.07达到了等效的观看次数（ENL），这将优于现有的最新OCT去斑方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
代理获取

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号