Algorithmic feature detection and statistical analysis in scanning probe microscopy data
Scanning Probe Microscopy has seen various advancements in recent years in terms of its imaging resolution, allowing researchers to image surfaces and particles on the atomic level. However, despite these advancements, the data acquisition processes undertaken in various fields are highly based on u...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/167521 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Scanning Probe Microscopy has seen various advancements in recent years in terms of its imaging resolution, allowing researchers to image surfaces and particles on the atomic level. However, despite these advancements, the data acquisition processes undertaken in various fields are highly based on user experience, leading to poor reproducibility of SPM results. Therefore, there is an increasing need to improve the replicability and accuracy of SPM results analysis through the use of various algorithms. While there are existing methods in use, they are not perfect. This project aims to develop a novel algorithm which utilises Topographical Prominence (TP) based on the OpenCV software package to automate and improve the reproducibility of data extraction and analysis of SPM data. The effectiveness of the algorithm is compared to several existing methods on their effectiveness of extracting the spatial distribution, count, and size statistics of SPM imaged nanoparticles using artificially simulated SPM images. Upon evaluation, it is observed that the TP method showed superior accuracy in analysing count and spatial distribution data as compared to other existing methods with potential of further work on optimising algorithm parameters and reducing computational load of the algorithm to extend its usage to higher definition SPM images. |
|---|