Surface-Defect-Detection

Why we included this project

Surface defect detection is one of those industrial vision problems where finding good training data is half the battle, and this repo exists to make that easier. It collects more than twenty public datasets, covering steel (NEU-CLS), PCB inspection (DeepPCB), rail surfaces, solar panels, and magnetic tiles, each with download links and a short rundown of what it contains. Alongside the data there is a chronologically ordered folder of key papers from 2017 on, plus a plain-language summary of the two issues that dominate the field: small sample counts and real-time inference. It is a curated index rather than runnable software, so treat it as a map for scouting training data and a reading list for understanding how defect detection is typically approached.