Preprocessing for Convolutional Neural Network (Pilot)

SUMMARY

“The quality of your model’s results is only as good as the quality of your data.” This course is designed to tackle one of the most crucial steps in developing reliable, high-performing CNN models for manufacturing—data preprocessing. When we look at images, subtle variations in lighting, angle, or noise may not distract us. But for a CNN tasked with spotting defects, these seemingly minor factors can make or break accuracy. How can we best prepare this data so our models can work effectively to ensure quality control and detect anomalies?

In this course, you will discover the preprocessing techniques that set up your data for success. From noise reduction to contrast enhancement, resizing, and data augmentation, each lesson focuses on equipping you with practical tools and best practices to handle visual data from sensors and cameras in industrial environments.

Whether you’re working with high-speed production line imagery or highly sensitive thermal and depth data, this course offers hands-on experience and actionable insights to make your data pipeline more robust. You’ll learn not only the how but the why of each technique, ensuring you leave with a solid foundation in preprocessing strategies that will elevate your CNN’s performance and make your quality control systems smarter and more reliable.

About The Author

Dilek Dustegor is a Professor of Computing Science at the University of Groningen in the Netherlands. She is interested in bridging the gaps between research, development and implementation using AI and automation. She is pursuing research about modeling, design and analysis of large scale / networked systems using IoT and ML techniques, with a special interest in smart city applications. She is a seasoned educator, and loves using the newest educational technologies for an enhanced learning experience.