Intelligent Autonomous UAS Inspection

By Stanley Armstrong, Samuel Frazee, Angello Parrolivelli, Guneet Kohli, Jason Ngu, Avi Peltz

• A1 - Decoder Component.ipynb
  • QR Code decoder component that utilizes PyZBar and OpenCV.
• A2 - Detection Component.ipynb
  • Attempt at creating an object-detection regression model. Though we were able to train it, we were unable to use it to predict on new test images.
• Component Integration.ipynb
  • Combines our decoder component and multi-label classifier to run our decoder on images that our classifier was able to pick up QR Codes on.
• Multi-label Classification.ipynb
  • Contains our multi-label classifier model which was successfully trained and able to output results on new test images.
• ONNX Model Export.ipynb
  • Outputs our classifier model to an ONNX format for cross-platform support.

Links

• Report
• Presentation slides