Skip to content
/ SETT2 Public

Create images of liquid crystal films with defects and their locations for training ML models.

Notifications You must be signed in to change notification settings

chowravc/SETT2

Repository files navigation

Note: Though named defect tracker, this module is SETT2

System for Enhancing simulated images, Training darkflow models, and Tracking ver. 2

Setup

The system has been tested using anaconda with python 3.6 on both linux and windows. to get requirements run:

pip install -r requirements.txt

All requirements should now be installed and SETT2 will be ready for usage.

Basic SETT2 usage

SETT2 is a module designed to be run by importing functions and run in an external python file. Create a python file in the same directory as this folder. Import sett with the following command.

import defectSimulation.sett2 as sett2

Run functions within sett2.

numImages = 10
imageDims = [250, 200]
maxDefects = 50
minDefects = 10
decrossMin = 50
decrossMax = 50

sett2.simulate(numImages, imageDims, maxDefects, minDefects, decrossMin, decrossMax)

Functionality

About

Create images of liquid crystal films with defects and their locations for training ML models.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published