TREC is a GUI application for video recording and playing. This application is especially interesting for new camera devices if they are supported by Video4Linux.
Two example screenshots are displayed below. The left screenshot displays TREC capturing the integrated camera of a laptop. The MLPG video codec with a resolution of 320x180 pixels at 15 frames per second (FPS) is selected (Many options are possible normally). The right screenshot displays TREC capturing a recorded video of an additive manufacturing process, Directed Energy Deposition (DED), to be specific. The video is played back at 90+ FPS.
Major information:
- 4 Threads; GUI, Display, Recording, Data acquisition for speed.
- Multiple resolutions and frame rates through Video4Linux integration
- Easy to expand for new devices i.e. Dinolite-edge etc.
- Build with Python,Qt, OpenCV and Numpy
- Live frame-rate and frame-rate capacity (%) based on thread capacity
Video4Linux device drivers are needed. To check run in terminal:
v4l2-ctlIf the package is not installed, install it with:
sudo apt install v4l-utilsAfter installation the visible devices can be checked with:
v4l2-ctl --list-devicesIf no devices are visible, the application will not be able to capture any. In that case only recorded videos can be played in the application.
Git has to be installed to clone:
sudo apt install gitClone the repository to current working directory
git clone https://github.com/gijsvanhoutum/trec.gitWe advise to install a new python virtual environment first with:
python3 -m venv venvActivate environment
source venv/bin/activateInstall all necessary Python packages with:
pip install -r /trec/requirements.txtTo run execute the following from the current working directory:
python3 /trec/trec/main.py- Make dynamic update of menu items possible in GUI. Currently, a newly created cannot be played back immediately. It requires a complete restart of the app.
- Complete comments for all classes.
- Currently Python Queues are used, which are thread-safe. All Mutex and locking is handled within the queue itself. A switch to QThreads is started (QMutex etc.) is implemented, but needs to be finished.