Autonomy-Interview-Challenge

---

If you don't have UV follow these instructions: https://docs.astral.sh/uv/getting-started/installation/

If you are on windows you may need to turn off smart app control

To run code do:

uv run main.py

---

Write all code in main.py

Challenge 0 (required):

Open interview_video.mp4 in openCV. Tutorial - https://learnopencv.com/read-write-and-display-a-video-using-opencv-cpp-python/

---

Challenge 1 (required):

Identify red LEDs in the video. Draw rotated bounding boxes around the red LEDs.

Hint - https://www.geeksforgeeks.org/python/splitting-and-merging-channels-with-python-opencv/

Suggested Steps:

• Show a video of extracted red channel
• Threshold red channel for the lights
• Contour binarized image
• Apply rotated rectangles to contours - https://www.geeksforgeeks.org/python/finding-minimum-enclosing-rectangle-in-opencv-python/

There are a couple of ways to remove small bits of noise, I'll let you figure that out 😃

---

Challenge 2 (Required if Autonomy and not embedded):

Use the bounding boxes from the previous challenge to find a position to the center of panel. PnP.py sample file contains an example of how to get the camera intrinsics, DO NOT EDIT or SUBMIT this file. Panel dimensions are included in the file. Print the position. A tutorial can be found here https://medium.com/@abdulhaq.ah/what-is-solvepnp-and-how-does-it-work-d9ac70823724

You may see strange values for X and Y positions, that is ok. Keep all of your units in cm.

---

Submition:

Upload main.py to the google form.