McMaster University, Department of Engineering Physics, Hamilton, L8S 4L7, Canada.
Sci Rep. 2018 Jan 24;8(1):1554. doi: 10.1038/s41598-017-18894-7.
Order classification is particularly important in photonics, optoelectronics, nanotechnology, biology, and biomedicine, as self-assembled and living systems tend to be ordered well but not perfectly. Engineering sets of experimental protocols that can accurately reproduce specific desired patterns can be a challenge when (dis)ordered outcomes look visually similar. Robust comparisons between similar samples, especially with limited data sets, need a finely tuned ensemble of accurate analysis tools. Here we introduce our numerical Mathematica package disLocate, a suite of tools to rapidly quantify the spatial structure of a two-dimensional dispersion of objects. The full range of tools available in disLocate give different insights into the quality and type of order present in a given dispersion, accessing the translational, orientational and entropic order. The utility of this package allows for researchers to extract the variation and confidence range within finite sets of data (single images) using different structure metrics to quantify local variation in disorder. Containing all metrics within one package allows for researchers to easily and rapidly extract many different parameters simultaneously, allowing robust conclusions to be drawn on the order of a given system. Quantifying the experimental trends which produce desired morphologies enables engineering of novel methods to direct self-assembly.
在光子学、光电学、纳米技术、生物学和生物医学中，有序分类尤为重要，因为自组装和生命系统往往有序但不完美。当（无序）结果看起来相似时，设计能够准确再现特定所需模式的实验方案集可能是一项挑战。在相似样本之间进行稳健的比较，特别是在数据集有限的情况下，需要一组经过微调的准确分析工具。在这里，我们介绍了我们的数值 Mathematica 包 disLocate，这是一套用于快速量化二维物体分散体空间结构的工具。disLocate 中提供的全套工具可以深入了解给定分散体中存在的质量和类型的有序性，访问平移、取向和熵有序性。该软件包的实用性允许研究人员使用不同的结构指标从有限的数据集中提取（单个图像）的变化和置信范围，以量化无序中的局部变化。将所有指标包含在一个包中，允许研究人员轻松快速地同时提取许多不同的参数，从而可以对给定系统的有序性得出稳健的结论。量化产生所需形态的实验趋势可以实现指导自组装的新方法的工程设计。
