Detect Defects on a Production Line for a Tier-1 Auto Supplier
Overview
What this challenge is about.
You receive 12,000 labelled grayscale part images (8,000 good, 4,000 defective across 6 defect types) at 2048x2048. Build a pipeline that does classical preprocessing (illumination correction, background subtraction) then a learned binary classifier. Optimize for a fixed false-accept rate of 0.05% and report the false-reject rate at that operating point. Add a per-defect-type confusion analysis. Deliver: pipeline code, a 4-page test report following ISO 9001 documentation conventions, and a 1-page recommendation for the shadow trial.
The Brief
What you'll do, and what you'll demonstrate.
Cut the line's false-reject rate by two-thirds without giving up on detection of real defects.
Earning criteria — what you'll demonstrate
- Apply classical preprocessing to industrial imagery
- Train a defect classifier at a fixed false-accept operating point
- Diagnose performance per defect type, not just in aggregate
- Communicate vision results in a quality-engineering format
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Computer Vision Engineer
Owning a defect-detection pipeline on a real production line is exactly the work CV engineers do at any Tier-1 supplier or industrial-vision vendor.
This challenge sharpens
- defect-detection
- industrial-vision
- image-preprocessing
Machine Learning Engineer
Operating-point selection at a fixed false-accept rate is the MLE skillset for any classification system with asymmetric error costs.
This challenge sharpens
- image-classification
- operating-point-analysis
- evaluation
Applied AI Scientist
Translating per-defect-type analysis into a shadow-trial recommendation is applied-AI work that drives plant-level decisions.
This challenge sharpens
- defect-detection
- operating-point-analysis
- evaluation