Development of AI-based Algorithms for Noodle Quality Perception and Assessment
KEYWORDKeywords
OBJECTIVE Objective
We aim to develop AI technology that perceives and assesses product quality across the five unit processes of noodle manufacturing using multi-sensor process data After validating this technology on a pilot line, we will transfer it to food companies, helping reduce raw material and process losses through the early detection of quality deviations. In addition, we will convert the quality-assessment know-how of skilled workers into data and train specialists in AI–food convergence.
INTRODUCTION Message from the Principal Investigator
Welcome.
I am Gyogwon Koo, Principal Investigator of this project.
The quality of the noodles on our tables is shaped through multiple manufacturing processes, yet assessing that quality has long depended on the experience of skilled workers.
Our research team was formed to carry this experience forward with data and AI, helping food manufacturers achieve consistent quality.
By combining DGIST’s AI research capabilities with the food process expertise of the Korea Food Research Institute (KFRI), we will build a collaborative model in which algorithms developed in the lab are validated with real noodle manufacturing data.
We will support our InnoCORE Fellows as they lead the entire research process, from data collection to field validation, and grow into convergence researchers who understand both AI and food engineering.
Through these efforts, we will lay the groundwork for AI transformation in food manufacturing
and do our utmost to ensure that our results are put to use in industry.
Thank you.
IMPACT Expected Impact
The quality perception technology and process data can be extended to other food processes with similar structures, such as confectionery and dried foods, while pilot-line demonstration cases can accelerate AI adoption by food companies.
By improving the consistency of food quality, we aim to strengthen food safety and consumer trust, and to expand the foundation for AI transformation (AX) in the food industry by supplying AI–food convergence talent to the field.
TALENT Ideal Talent Profile
Data-Grounded : A field-oriented researcher who understands sensor principles and process conditions, carries out every step from data measurement and alignment to performance validation, and derives research questions from real-world problems
Domain-Bridging : A convergence-minded researcher who combines AI modeling skills with food engineering and process knowledge to solve problems spanning heterogeneous data such as spectra, time series, and images
Trustworthy & Collaborative : A collaborative researcher who treats reproducible experiments and honest performance reporting as the foundation of research, and who works with research institutes, companies, and shop-floor operators to turn research outcomes into technology used in the field