A preprocessing and uploading app for barn data in an edge computing setting¶
A screenshot of the preprocessing and upload app
🎯 Project objective¶
Enable biologists to upload data to the remote storage from the barn using secure and fast protocols.
Required to run preprocessing steps in local before sending to remote storage in an edge computing setting.
📋 Project description¶
A project that builds upon the McGill barn video uploads app.
This desktop app form preprocesses the video data on-site (a barn) before uploading video files to the WELL-E file storage server. On-site preprocessing consists of multiple operations such as anonymization of barn staff, video trimming, video encryption, etc. to reduce the size of files transferred and to keep security optimal.
🎨 Design & implementation decisions¶
Python3 GUI
Python3 classes for preprocessing
System automations for Globus endpoint transfer to storage server
🧾 Key takeaway¶
GUI improvement with Flet
ML preprocessing of videos
Globus endpoint for efficient and secure file transfer
👨💻 Contribution:¶
Design & build the GUI
Database design and implementation
Backend usage of code from other private packages
Automate preprocessing operations depending on the information filled in the form
Automate metadata gathering and remote database update depending on the information filled in the form
🛠 Tools:¶
Python
Flet
SQLite3
PyTorch
YOLOv7
OpenCV
Pandas
Seaborn
Globus endpoints