https://github.com/jose-elias-alvarez/flybox-scanner
https://github.com/jose-elias-alvarez/flybox-scanner
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/jose-elias-alvarez/flybox-scanner
- Owner: jose-elias-alvarez
- Created: 2023-08-08T16:48:35.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-01-09T03:59:06.000Z (almost 2 years ago)
- Last Synced: 2025-07-19T16:43:32.919Z (3 months ago)
- Language: Python
- Size: 4.36 MB
- Stars: 0
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# flybox-scanner
## Setup
First, make sure [Anaconda](https://www.anaconda.com) is installed.
Then, navigate to the project directory and run the following commands:
```sh
conda create --prefix .conda python=3.11
conda activate ./.conda
```
This will create a new environment in the current directory and activate it.
You'll then need to install the project's dependencies:
```sh
conda install --file requirements.txt
```
### Direnv (optional)
You can streamline activation by setting up [direnv](https://direnv.net/), which will automatically activate and deactivate the environment as needed.
### Start
If needed, activate the environment:
```sh
conda activate ./.conda
```
Then, run the script:
```sh
python main.py
```
### Uninstall
To uninstall the environment, run the following command:
```sh
conda env remove --name flybox-scanner
```
## Parameters
### Motion Detection
To tune motion detection parameters, run the following command:
```sh
python main.py --tuning=motion
```
Once you've found a set of parameters that works well, you can save them to `settings.json` using the `Save` button in the UI. The parameters will then be loaded automatically on subsequent runs.
### Other Parameters
Other parameters are currently hard-coded into the following Python source files and must be edited manually:
- `main.py`: controls source and window parameters
### Detection
- `detection/border.py`: controls the detection of borders in frames (used to "crop" the flybox background)
- `detection/grids.py`: controls the detection of wells
### Handlers
- `handlers/debug.py`: controls the display of debug info in the recording window (e.g. wells, flies)
- `handlers/file_interval.py`: controls the output of data into a file at a given interval
- `handlers/frame.py`: controls the handling of detected motion and its conversion into motion events
## Development
### Architecture
UI logic lives in `components/`. The main UI entrypoint is `components/root_window.py`, which is responsible for the main event loop and for mounting and unmounting "state canvases".
Each state canvas is responsible for a specific UI state, as defined in `components/state_manager.py`:
- `components/idle_canvas.py`: initial state, displays a source preview and state transition buttons
- `components/select_webcam_canvas.py`: active when selecting a webcam source
- `components/scan_canvas.py`: active when "scanning" (running grid detection on) the source, allows transitioning to the recording state
- `components/record_canvas.py`: active when recording, shows a (hideable) preview
### Videos
Note that video output is not yet optimized and may differ from webcam output.
- [Fly.mp4](https://drive.google.com/file/d/1q6RSJJIWKrrxvLqLVuanaOrmj1ull6yN/view?usp=share_link)
- [DoubleFly.mp4](https://drive.google.com/file/d/1jw3vVR3u8bQfJR4toDuorEAYqEPXu1Qc/view?usp=share_link)
### Tests
Tests are written using `unittest` and live in the `tests/` directory. To run them, use the following command:
```sh
python -m unittest discover tests/
```
### VS Code
#### Tasks
The repository contains a `Run` task to run `main.py`. To use it, open the command palette and select `Tasks: Run Task`. Then, select `Run`.
#### Interpreter
Upon creating the new environment, VS Code should prompt you to use it for the current workspace. If not, use the `Python: Select Interpreter` command to select the interpreter at `./env/bin/python.`