Objective Visual Inspection Measurement Result in Counterfeit IC Detection Using Artificial Intelligence and Computer Vision

Chih-Yun Pai and Tom Zheng [Global ETS]

Abstract: 

While various methods exist to verify IC authenticity, visual inspection remains essential. Standards like AS6171 define its conceptual framework but lack objective measurement tools. As a result, visual inspection frequently depends on the inspector’s subjective judgment, with features such as surface texture, mold-pin cavities, lead contamination, sanding marks, and label authenticity assessed by eye. This subjectivity leads to variability, limits traceability and scalability, and results in inconsistent ‘pass/fail’ decisions, complicating quality control and audits.
To address this subjectivity, analytical modules were developed to quantify key aspects of visual assessment. Among the indicators examined, two are highlighted in this study: surface texture consistency analysis using an image quality assessment (IQA) model and font analysis based on OCR and clustering. This study shows how visual attributes can be parameterized into quantitative scores to improve inspection consistency.
The algorithm and software developed by Global ETS have undergone extensive research and development to overcome these challenges and now serve as an AI- assisted analytical tool that complements domain expertise. Its machine learning models are trained via a semi-supervised learning approach on a curated in-house library of authenticated and counterfeit IC images. For each device under test, it delivers an annotated report with clear anomaly scores, highlights regions of interest, and provides concise summaries to support decisions.

Biography: 

Dr. Chih-Yun Pai is a manager and research scientist at Global ETS. He leads the company’s AI and Computer Vision research projects. Dr. Pai joined Global ETS in 2021. At Global ETS, he conducts research on using AI and computer vision to assist and automate the process of visual inspection. He is now leading the BlueOptiView research team to develop an AI-assistive visual inspection service.
Dr. Pai received his Ph.D. Degree in Computer Science and Engineering from the University of South Florida in 2021 and his MS. Degree in Computer Science from the same university in 2016. Dr. Pai is the author of six peer-reviewed papers in the field of machine learning and medical image and holds one patent and one pending application.

Tom Zheng works as a Senior Software Engineer at Global Electronics Testing Services (GETS). A 2021 graduate of the University of South Florida with a B.S in computer science. Since joining GETS in 2021, he led and architected Lotus, which is the lab’s central test management and tracking platform. The platform orchestrates testing sequences and automation from test stations to tune operations in the semiconductor testing laboratories.
Tom collaborated with BlueOptiView’s research team to develop and launch an AI assistive visual inspection service. He designed and deployed a scalable cloud infrastructure and pipeline to support engineers from manual reviews and accelerating supervised training cycles.

Dr. Diganta Das

For more information or questions regarding the technical program (including Professional Development Courses), contact the Conference Chair, Dr. Diganta Das

Kristin Nafstad

For more information or questions regarding event logistics, exhibitions, and sponsorship, contact Kristin Nafstad.