Objective: This module outlines the basic principles for automated optical inspection (AOI) of surface-mount devices on printed circuit board (PCB) images. Sub-topics include quantifying image quality, key features during image acquisition, importance of data collection for machine learning trends, and challenges encountered during image annotation. On course completion, students should have an enhanced understanding of the data analysis challenges during PCB AOI and state-of-the-art trends.
Target Audience: Government officers, Scientists
Prerequisite Knowledge and Skills:
- programming knowledge: python
- IC design flow steps
Resources Provided at the Training | Deliverables:
- Detailed description of set-ups used in training
- A video demo of the module
Learning Outcome: This module covered the basics of AOI, formats of image annotation and ML data preparation, model loading, and segmentation mask generation. While the data used in the training only came from one image, the demonstrated approach can scale with additional images to produce more viable models.