Thompson Russell E, Larson Daniel R, Webb Watt W
Cornell University, School of Applied and Engineering Physics, Ithaca, New York 14853, USA.
Biophys J. 2002 May;82(5):2775-83. doi: 10.1016/S0006-3495(02)75618-X.
Calculation of the centroid of the images of individual fluorescent particles and molecules allows localization and tracking in light microscopes to a precision about an order of magnitude greater than the microscope resolution. The factors that limit the precision of these techniques are examined and a simple equation derived that describes the precision of localization over a wide range of conditions. In addition, a localization algorithm motivated from least-squares fitting theory is constructed and tested both on image stacks of 30-nm fluorescent beads and on computer-generated images (Monte Carlo simulations). Results from the algorithm show good agreement with the derived precision equation for both the simulations and actual images. The availability of a simple equation to describe localization precision helps investigators both in assessing the quality of an experimental apparatus and in directing attention to the factors that limit further improvement. The precision of localization scales as the inverse square root of the number of photons in the spot for the shot noise limited case and as the inverse of the number of photons for the background noise limited case. The optimal image magnification depends on the expected number of photons and background noise, but, for most cases of interest, the pixel size should be about equal to the standard deviation of the point spread function.
计算单个荧光颗粒和分子图像的质心，可在光学显微镜中实现定位和跟踪，其精度比显微镜分辨率高约一个数量级。研究了限制这些技术精度的因素，并推导了一个简单方程，该方程描述了在广泛条件下的定位精度。此外，构建了一种基于最小二乘法拟合理论的定位算法，并在30纳米荧光珠的图像堆栈和计算机生成的图像（蒙特卡罗模拟）上进行了测试。该算法的结果与模拟图像和实际图像的推导精度方程都显示出良好的一致性。一个描述定位精度的简单方程的可用性，有助于研究人员评估实验设备的质量，并将注意力指向限制进一步改进的因素。在散粒噪声受限的情况下，定位精度与光斑中光子数的平方根成反比；在背景噪声受限的情况下，与光子数成反比。最佳图像放大倍数取决于预期的光子数和背景噪声，但在大多数感兴趣的情况下，像素大小应约等于点扩散函数的标准差。
