Beamline I15-1 / I15-J Robot Camera System aimed towards identifying the presence of sample holder lids. This in turn prevents any attempts for samples to be collected while a lid is attached to a sample puck. This is done using OpenCV and FastAPI to provide calls for position states, both overall ["result"] and individual ["position"] as well as an annotated image ["image"] to cross-check the accuracy of the image.
Service Workflow: 0. Calibrate the file to dewarp the image to collect the stage.
- Collect Image from stream
- Dewarp Image
- Apply Adaptive Thresholding to the image
- Apply Gaussian Blur to image
- Identify and sort positions using Hough Transforms
- Identify state of positions ["None", "Puck", "Lid"]
- Return FastAPI response - ["result"], ["position"], ["image"]
Documentation: User Guide: https://confluence.diamond.ac.uk/display/SSCC/Documentation%3A+User+Guide Developer Guide: https://confluence.diamond.ac.uk/display/SSCC/Documentation%3A+Developer%27s+Guide