Skip to content

This project aims to extract data from thin section images, using it to return the number of minerals in the image, the modes of chosen minerals, and their cumulative size.

Notifications You must be signed in to change notification settings

kylewjang/Thin-Section-Processing

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

25 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Thin-Section-Processing-WIP

Eyeballing mineral modes can be difficult sometimes, especially when there are many different minerals in a thin section, and I wanted to try to make it easier.

This script takes any colour image as an input, so both PPL and XPL views should work.

Above are the original scripts created on Pycharm (.py) and the Jupyter Notebooks (.ipynb)

A thin section of Andesite found on Google images was used as an example.

Method 1: Phases - Created a new image by allocating different values corresponding to the pixel values of the minima of the filtered image histogram. For example, the first phase is for values between 0 and the first minima. The limited number of different pixel values allows for easier classification of colours, if desired.

Method 2: Using the unphased filtered data

The accuracy of the methods is decent, and each method has its perks. However, there is likely a more accurate method that I couldn't seem to come up with within the limited time I had.

Please note that the logic behind this project is not perfect. I am aware that colour alone is not sufficient for mineral identification, but it can be a distinctive feature. The main purpose of this project was to find a reasonably accurate mineral modes in an image, not neccesarily to return the precise number of distinct minerals in the image.

Feel free to use these resources in any way you'd like!

About

This project aims to extract data from thin section images, using it to return the number of minerals in the image, the modes of chosen minerals, and their cumulative size.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published