It is commonly hard to take image using CMOS (or CCD) sensor during taking MRI (magnetic resonance imaging) signal due to the magnetic compatibility or electromagnetic interference. For that, we bought commercialized MR compatible camera and have developed this software to control the devices conveniently. This tool implement the keypoint detection model to automatically measure the pupil size in real time. If you want to detect keypoint from another type object, refer this site and train for your data
Copy below files from here and paste to lib directory
TIS_UDSHL11_x64.dll
tisgrabber_x64.dll
Refer to the conda envrionment file for installing python packages
CPU : AMD Ryzen 9 5900X
GPU : NVIDIA GeForceRTX 3080Ti
memory : 64 GB
OS : windows11
Virtual env : Conda
TTL generator : Master9
We used commercial MR compatible camera(12M-i, MRC), filter box and TTL recevier (Adventech) bought from MRC.
I designed this software to recognize the TTL signal that Vpp=5V, Vmax=5V, Vmin=0 and duration > 200ms
Run FlyingSesame.py file in terminal
python FlyingSesame.py
[1] https://github.com/TheImagingSource/IC-Imaging-Control-Samples
[2] https://github.com/DeepLabCut/DeepLabCut