Objective: Printed Circuit Boards (PCBs) are the carrier of many passive and active components for consumer, corporate, and military applications. Securing these devices from tampering through device verification is critical to national security and consumer safety. This requires non-destructive process such as X-ray imaging which requires advanced and expensive machines. In this module, users will learn how to gather X-ray data, develop image processing techniques, and apply it to various counterfeit PCB and IC components. There are many components integrated onto a PCB, while the PCB contains many points of failure or medication, and this requires a robust platform for various academic and corporate users to gather training data for using artificial intelligence detection systems. This manual labelling of the vias and trace’s specific location, size, direction, and width of features are stored and can be recalled once needed to train different ML or deep learning (DL) algorithms. Users will learn how to engineer which features from the X-ray data are most critical for automating A.I. or DL systems.
Target Audience: Government officers, Scientists
Prerequisite Knowledge and Skills:
- programming knowledge: python
Resources Provided at the Training | Deliverables:
- Detailed description of set-ups used in training
- A video demo of the module
Learning Outcome: In this module, trainees will learn how to gather X-ray data, develop image processing techniques, and apply it to various counterfeit PCB and IC components.