Britton Chance Center for Biomedical Photonics, Wuhan National Laboratory for Optoelectronics-Huazhong University of Science and Technology, Wuhan 430074, China.
Opt Lett. 2013 Jun 1;38(11):1769-71. doi: 10.1364/OL.38.001769.
We present a camera embedded data processing method for localization microscopy (LM) with faster detectors such as scientific complementary metal-oxide semiconductor (sCMOS) cameras. Based on the natural sparsity of single molecule images, this method utilizes the field programmable gate array chip inside a camera to identify and export only the regions containing active molecules instead of raw data. Through numerical simulation and experimental analysis, we found that this method can greatly reduce data volume (<10%) with negligible loss of useful information (<0.2%) at molecular densities <0.2 molecules/μm(2), thus significantly reducing the challenges of data transfer, storage, and analysis in LM.
我们提出了一种适用于具有更快探测器（如科学互补金属氧化物半导体（sCMOS）相机）的定位显微镜（LM）的相机嵌入式数据处理方法。基于单分子图像的自然稀疏性，该方法利用相机内部的现场可编程门阵列芯片来识别和仅输出包含活性分子的区域，而不是原始数据。通过数值模拟和实验分析，我们发现该方法可以大大减少数据量（<10%），同时在分子密度<0.2 分子/μm(2)时，有用信息的损失可忽略不计（<0.2%），从而显著降低了 LM 中数据传输、存储和分析的挑战。
