(The content below # Snowball2 is all out of date, the following information is correct until the next closing statement)
In each of the folders named with a single letter of the alphabet (A, B, C,...) and each the begins with "Run" following by a numeric index, are videos and plots of results of image analysis. These videos have 4 frames, from left to right they are:
- The original video
- The video after numerous filters are applied
- The binary mask generated by OpenCV (not as straightforward as a single function call)
- Hough Circles, overlayed with one another to show historic and current tracking of multiple snowballs. The numbers are an index, number of snowballs counted, and the current frame number (out of ~201). The appropriate time density is about 20 frames per second, but varies between 7.5 and 30 frames per second. While the analysis procedure is purely based on evaluating adjacent frames as equidistant, visualizations have been performed that use the timestamp of each frame to match the relative real-time elapsed between frames. Such a video may or may not be on this repository (it has been a year, I simply do not know).
If you would like to see the best illustration of what has been done here, I suggest TestVideo.avi is a good clip to view.
(End. All content below this point is out of date.)
Scott Schwartz software used: - External code editor: VScode - Python 3.9.7 64-bit (windows store) - R2021a ("academic use") - (refers to matlab?)ds - GitHub Desktop - Extensions: - Matlab Extension Pack: https://marketplace.visualstudio.com/items?itemName=bat67.matlab-extension-pack - vscode-pdf: https://marketplace.visualstudio.com/items?itemName=tomoki1207.pdf - vscode-tiff: https://marketplace.visualstudio.com/items?itemName=ucodkr.tiff-preview - Various py and jupyter extensions - "jupyter" extension, "jupyter keymap" extension, "Pylance" extension, "Python" extension - all help with viewing code within visual studio code in a more visually friendly manner while editing and viewing it.
must have pip: https://pip.pypa.io/en/stable/installation/
enter these commands in the cmd prompt:
pip install numpy pip install matplotlib pip install opencv-python
Extensions: image resizer (bmp viewing) python (has multiple extensions) opencv snippets simply_view_image_for_python_opecv_debugging vscode-pdf vscode-tiff
File size management: in order to make the repository not excessively large a solution for keeping the cropped images is currently being developed using a second repo
BE SURE TO VIEW whichRuns.txt https://github.com/schwas3/Snowball2/blob/main/Bar%20and%20Hist%20Figs/whichRuns.txt