中国激光, 2020, 47 (2): 0207024, 网络出版: 2020-02-21
基于噪声校正主成分分析的压缩感知STORM超分辨图像重构 下载: 1322次
Compressed Sensing STORM Super-Resolution Image Reconstruction Based on Noise Correction-Principal Component Analysis Preprocessing Algorithm
生物光学 随机光学重构显微 主成分分析 降噪算法 超分辨光学成像 biotechnology stochastic optical reconstruction microscopy principal component analysis denoising algorithm super-resolution optical imaging
摘要
随机光学重构显微(STORM)的时间和空间分辨率相互制约,难以实现活细胞的超分辨成像,且超分辨图像的后处理分析与重构算法对图像质量也有非常重要的影响。基于此,针对高密度标记与高采样率所导致的单帧图像中光斑重叠及过多的背景噪声,提出一种用于单分子定位显微成像的新型噪声校正主成分分析(NC-PCA)方法,对单分子定位显微成像采集的图像进行预处理后再进行定位重构,提高了现有定位方法的定位精度,同时还实现了重叠分子的区分定位,从而提高了生物样品的标记密度,改善了超分辨成像的时间分辨率,可为活细胞单分子定位成像提供技术支持。
Abstract
The low temporal resolution of stochastic optical reconstruction microscopy (STORM) limits its ability to observe dynamic events in live cells. Further, the post-processing analysis and reconstruction algorithms have an important effect on super-resolution images. In this study, we report a new noise-correction principal component analysis method for single-molecule localization microscopy against fluorescent spot overlapping and excessive background noise in a single frame of images owing to high-density labeling and high camera-sampling frequency. The proposed method can improve the positioning accuracy of existing localization methods by pre-processing the raw images acquired by the single molecule localization microscopy before reconstruction. In addition, this method can accurately distinguish the overlapping molecules. Therefore, it is suitable for samples exhibiting a high fluorophore density. Thus, the proposed method improves the temporal resolution of super-resolution imaging, providing a powerful technical support for the STORM imaging of live cells.
潘文慧, 陈秉灵, 张建国, 顾振宇, 熊佳, 张丹, 杨志刚, 屈军乐. 基于噪声校正主成分分析的压缩感知STORM超分辨图像重构[J]. 中国激光, 2020, 47(2): 0207024. Pan Wenhui, Chen Bingling, Zhang Jianguo, Gu Zhenyu, Xiong Jia, Zhang Dan, Yang Zhigang, Qu Junle. Compressed Sensing STORM Super-Resolution Image Reconstruction Based on Noise Correction-Principal Component Analysis Preprocessing Algorithm[J]. Chinese Journal of Lasers, 2020, 47(2): 0207024.