Improving semiconductor visual inspection processes using deep learning techniques

Abstract

I completed my master’s end-of-study internship at STMicroelectronics as a Data Scientist. The company specializes in the design and manufacturing of semiconductors. I joined the Data Science and Analytics team, a young international team whose main objective is to address all data science issues to improve performance in terms of manufacturing and process quality. During this internship, I worked on two major computer vision projects. The first project involved developing an image classification algorithm to improve the quality of the automatic inspection process of flip chips on the A49 assembly line at the backend site in Malta. For this project, I also developed a Spotfire dashboard to provide engineers with a real-time monitoring tool. The second project was a research and development mission. The objective was to develop a Python library for automatic multi-label image classification. The idea was to provide a flexible tool for Data Scientists that would save considerable time. This internship allowed me to develop and strengthen my skills in data science, particularly in image analysis. I also learned about best practices for teamwork, communication, and collaboration. In this sense, this experience was extremely enriching.

Date of Award	17 Oct 2023
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Christian Schluter (Supervisor)