Chonnam National University Joint Research Team Develops AI-Based Defective Product Detection Model

[Asia Economy Honam Reporting Headquarters Reporter Lee Gwan-woo] A research team at Chonnam National University has attracted attention by developing a model that identifies defective products using artificial intelligence (AI).

On the 10th, according to Chonnam National University, Professor Kim Jin-young's Intelligent Electronics Laboratory team and Professor Kim Jin-sul's Smart Mobile/Media Computing Laboratory team jointly developed an 'AI-based model for distinguishing good and defective products' that can identify defective products occurring during the manufacturing process on production sites.

The model achieved an accuracy of as high as 99.2% in testing and verification, which is expected to contribute not only to increasing production automation rates but also to accelerating the realization of smart factories.

The research team presented these results at the '9th International Conference on Smart Media and Applications (SMA 2020)' held in September, organized by the Smart Media Society, and received the 'Best Paper Silver' award.

Based on this research, the team plans to continue studies to improve algorithms for enhancing the accuracy of deep learning models and to integrate them with PLC modules used in production sites.

Meanwhile, the research team is concurrently conducting the project 'Development of AI-based Production Site PLC Modules and Control Programs,' led by the Ministry of Science and ICT and the National IT Industry Promotion Agency, with cooperation from Gwangju Metropolitan City and the Gwangju Information & Culture Industry Promotion Agency.

This content was produced with the assistance of AI translation services.