Generalizable Framework for Algorithmic Interpretation of Thin Film Morphologies in Scanning Probe Images

Abstract

We describe an open-source and widely adaptable Python library that recognizes morphological features and domains in images collected via scanning probe microscopy. π-Conjugated polymers (CPs) are ideal for evaluating the Materials Morphology Python (m2py) library because of their wide range of morphologies and feature sizes. Using thin films of nanostructured CPs, we demonstrate the functionality of a general m2py workflow. We apply numerical methods to enhance the signals collected by the scanning probe, followed by Principal Component Analysis (PCA) to reduce the dimensionality of the data. Then, a Gaussian Mixture Model segments every pixel in the image into phases, which have similar material-property signals. Finally, the phase-labeled pixels are grouped and labeled as morphological domains using either connected components labeling or persistence watershed segmentation. These tools are adaptable to any scanning probe measurement, so the labels that m2py generates will allow researchers to individually address and analyze the identified domains in the image. This level of control, allows one to describe the morphology of the system using quantitative and statistical descriptors such as the size, distribution, and shape of the domains. Such descriptors will enable researchers to quantitatively track and compare differences within and between samples.

Kelvin Probe Microscopy Investigation of Poly-Octylthiophene Aggregates

“…The higher lamellar structures ( Figure 4 ) have flat layers of approximately constant height and up to five layers (in all the studied structures in HOPG). In order to study all the data measured by SFM—topography, frequency shift and surface potential (KPFM data)—in more detail, we use the m2py library [ 50 ], which essentially allows us to detect “different regions” using a segmentation procedure explained in more detail elsewhere [ 51 ]. When this segmentation procedure is applied to the topography image, it clearly recognizes the different layers of the lamellar structure, as shown in Figure 4 d, where each color codes the different regions recognized by the algorithm (including HOPG and the inter-lamellar structure), and in particular the different layers of the lamellar structure.…”

“…Data analysis was carried out using the open-source Materials Morphology Python m2py library [ 50 ]. This versatile and widely adaptable code combines computer vision and machine learning techniques to recognize different sample regions as well as different material properties depending on the input channel used for the segmentation process, as explained in detail in [ 51 ]. Once the image is segmented in different regions, the properties of each region obtained from other acquisition channels can be analyzed independently.…”

Recognition Method of Digital Meter Readings in Substation Based on Connected Domain Analysis Algorithm

“…In the image processing, the domain [39] can be added to the mark for processing. The thesis proposes an analysis and judgment algorithm to solve the problem of whether there is a decimal point in the digital area of the dial.…”

Data-driven materials research enabled by natural language processing and information extraction

“…Tatum et al have also recently reported an image-recognition method that provides quantitative analysis of particles appearing specifically in images created by scanning probe microscopy (SPM) techniques, such as STM and AFM. 134 Particles are first detected using feature selection that is enabled by principal-component analysis (PCA); this clusters all data channels into the key representative structure of the image-based information. These clustered data are, then, classified using a Gaussian mixture model (GMM), which segments each pixel into distinct material phases; in the case study, the phases are structural domains of a polymer blend.…”