谢巧雪,马宗庆,祝连庆,朱 疆.基于结构相似约束生成对抗网络的视网膜 OCT 图像去噪算法[J].电子测量与仪器学报,2023,37(3):11-20 |
基于结构相似约束生成对抗网络的视网膜 OCT 图像去噪算法 |
Retinal OCT image denoising based on structural similarity constrained generative adversarial network |
|
DOI: |
中文关键词: 光学相干断层成像 视网膜 图像去噪 生成对抗网络 结构相似损失 |
英文关键词:optical coherence tomography retina image denoising generative adversarial network structural similarity loss |
基金项目:国家自然科学基金( 61975019)、北京市教育委员会科技计划重点项目( KZ202011232050)、北京信息科技大学校科研基金(2021XJJ10)项目资助 |
|
Author | Institution |
Xie Qiaoxue | 1. Key Laboratory of the Ministry of Education for Optoelectronic Measurement Technology and Instrument, Beijing Information Science and Technology University,2. Beijing Laboratory of Biomedical Testing Technology and Instruments,Beijing Information Science and Technology University |
Ma Zongqing | 1. Key Laboratory of the Ministry of Education for Optoelectronic Measurement Technology and Instrument, Beijing Information Science and Technology University,2. Beijing Laboratory of Biomedical Testing Technology and Instruments,Beijing Information Science and Technology University |
Zhu Lianqing | 1. Key Laboratory of the Ministry of Education for Optoelectronic Measurement Technology and Instrument, Beijing Information Science and Technology University,2. Beijing Laboratory of Biomedical Testing Technology and Instruments,Beijing Information Science and Technology University |
Zhu Jiang | 1. Key Laboratory of the Ministry of Education for Optoelectronic Measurement Technology and Instrument, Beijing Information Science and Technology University,2. Beijing Laboratory of Biomedical Testing Technology and Instruments,Beijing Information Science and Technology University |
|
摘要点击次数: 608 |
全文下载次数: 967 |
中文摘要: |
光学相干断层扫描(OCT)图像中存在的散斑噪声会掩盖视网膜重要的形态学细节,妨碍视网膜病变的观察和临床诊
断。 提出了一种基于结构相似约束生成对抗网络的视网膜 OCT 图像去噪算法,基于残差策略改进生成对抗网络模型结构,并
融合结构相似性损失约束模型优化,实现散斑噪声抑制,同时增强对视网膜结构细节的保留。 在杜克大学发布的 SD-OCT 公开
数据集上的实验表明,所提算法的峰值信噪比和边缘保持指数分别为 28. 08 和 0. 960,优于所对比的其他去噪方法,且适用于其
他来自 A2A SD-OCT 研究的公开数据集。 |
英文摘要: |
Speckle noise in optical coherence tomography (OCT) images obscures important morphological details, and hinders the
clinical observation and diagnosis of retinal lesions. A structural similarity constrained generative adversarial network ( SSGAN) is
proposed for retinal OCT image denoising. The proposed SSGAN utilizes the residual strategy to improve the structure of original
generative adversarial network, and incorporates the structural similarity index measure loss into the objective function to achieve more
structural constrains while suppressing speckle noise. The experiments on the SD-OCT public dataset published by Duke University show
that the peak signal-to-noise ratio and edge preserve index of the proposed method are 28. 08 and 0. 960 respectively, outperforming the
other denoising comparison methods. Further experiments demonstrate that the proposed method can be easily applied to other public
datasets from the A2A SD-OCT study. |
查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|