Microfluidic Droplet Ejection Computer Vision
The Objective
To analyze thousands of microfluidic droplet ejection videos for the Medically Advanced Devices Laboratory (MAD Lab) at UC San Diego in an accurate, consistent, and time effective way.
Personal Achievements
- Wrote computer code using ImageJ and MATLAB to automatically analyze thousands of videos of ejected fluid droplets using machine vision.
- Compiled excel sheet of collected data to aid in developing underlying theory behind ultrasonic fluid droplet ejection.
- Devised a way to queue droplets automatically before being ejected for potential use in a new type of 3D printing technology.
The Outcome
I learned how to use batch processing in a very powerful open source image processing software, ImageJ, and was also able to flex my MATLAB skills a bit. The useful data was extracted from thousands of slow motion video files in a tiny fraction of the time it would have taken without my code. I am now starting the process of adapting the code to the projects of other graduate students in the same laboratory at UC San Diego.
A little snippet of an altered version of the code that runs entirely in MATLAB and uses their image processing toolkit for better visualization purposes. I will be publishing this soon too!
A Video Explanation of the Final Product
Examples of Tracking In Matlab
Generated Text File With Data!
Some Other Ways I was Able to Visualize Data
