This lab works on phones, but writing code is much easier on a laptop.
Perception · Image Processing for Robotics · P2-L2 · Beginner
A robot finds dark parts on a light conveyor belt. It works in the afternoon. In the morning, and when a cloud passes, it sees parts everywhere.
You will see why a fixed number cannot work, make the detector follow the overall light, and then make it follow light that changes across the image.
You have passed every stage. Now connect it back to the theory and move on:
Plain Python with math (and slamkit for small matrix helpers: matmul, transpose, solve, inverse, identity). Images, point clouds and detections arrive as lists inside the dictionary D; an image is a list of rows, each row a list of pixels. There is no NumPy or OpenCV: you write the algorithm yourself, which is the point of the lab. Print with print().
Stage 1 is a controlled experiment with two light levels. Stage 2 asks you to write a global normalisation. Stage 3 asks you to write a local adaptive threshold. Each is graded on hidden images with different light levels and gradients; the simulator keeps the true object positions private.
Educational simplification: simulated 40 by 28 grey images of dark rectangles on a light floor, scaled by a light level, with an optional left-to-right shading gradient and Gaussian pixel noise. Real lighting also produces shadows with hard edges, highlights and colour casts. All images in this lab are simulated.
Get notified when new robotics labs are released. One email per new lab, nothing else.