Project Title

Automated Defect Classification 2.0

Description

The Automated Defect Classification (ADC) 2.0 project aims to improve the accuracy and efficiency of defect detection across multiple inspection platforms, including ATI Wind, Dragonfly, and Sam Scan. The primary goal is to train advanced ADC models and validate their performance to ensure consistent and reliable defect classification in the production line. This involves using brand new machine learning techniques to automate the identification and categorization of defects, thereby reducing human error and improving overall inspection quality.

Scope

• Engage with multiple teams and departments: collaborate with Engineering, IT, QA, and SMAI to gather diverse expertise and insights for model training.
• Use RDAViewer for data verification: apply RDAViewer to verify the accuracy of defect data and ensure the integrity of the training dataset.
• Data collection a...