JoshTheEngineer | How To Take Pictures Like NASA: DIY Background Oriented Schlieren @JoshTheEngineer | Uploaded October 2019 | Updated October 2024, 2 hours ago.
Want to take schlieren images or videos at home, but don't have access to a nice mirror? Or you've mastered the conventional schlieren technique and want to try your hand at something a little different? Or like the title mentions, do you want to take schlieren images like NASA? In this video, I'll walk you through every step you need to get started with Background Oriented Schlieren (BOS) experiments in your own home. Everything can be done with programs that are free to use. I also include my specially-written Python and MATLAB GUI programs for you to use.

UPDATED: Please use the second version (v2) of my ImageJ macro so you don't run into any issues. Here is my update YouTube video: youtube.com/watch?v=GXapOtBf52o

==== TIME-STAMPS ====
00:00 - Introduction
01:25 - Outline
02:32 - Programs you will need
03:08 - FFmpeg
05:28 - ImageJ/Fiji
13:42 - ImageJ macros
16:39 - Python code (images)
23:38 - MATLAB code (images and videos)
36:05 - DIY at home

==== NOTES ====
► This is a long video, I know. But I would rather upload a long, comprehensive video than skimp on the details. But if you're a regular to my channel, you probably know that already.
► My programs are not completely fool-proof. If you do something horribly wrong, you might get an error. But, if you use it the way I walk you through in the video, you shouldn't have any issues. If you do get an error, it might be best to just restart the program.
► I will probably be adding the video capability to the Python GUI at some point. Or better yet, you can fork the repository on GitHub and practice your Python coding by adding it in yourself.
► Not every video you download from YouTube will work when trying to convert it like I did with FFmpeg. Most of the ones I tried worked, but if yours doesn't, I'm not the best person to ask how to fix it. Google is helpful here.
► The NASA data obviously looks really nice. My processing of their data doesn't look as nice. There are many reasons for this, including (but not limited to) the high quality of their raw images, averaging multiple images together, using a more sophisticated code, and using a different analysis technique (optical flow).

==== RELEVANT VIDEOS ====
Schlieren YouTube Playlist
► youtube.com/watch?v=GEImPtOM-d4&list=PLxT-itJ3HGuWift1-NXVhSCzGd9uqDJv1

==== RELEVANT LINKS ====
My website with code and PDF document
► joshtheengineer.com/2019/10/20/how-to-take-pictures-like-nasa-using-background-oriented-schlieren-bos

My GitHub with code
► github.com/jte0419/Background_Oriented_Schlieren

==== REFERENCES ====
Python's "normxcorr2" function from GitHub user Sabrewarrior
→ github.com/Sabrewarrior/normxcorr2-python

NASA AirBOS Shock Wave Video
→ youtube.com/watch?v=443RLEnu_UI

Find all the rest of my references in my PDF document