A CNU research team has developed a model that identifies defective products with artificial intelligence (AI).

The research team of Professor Kim Jin-young of the CNU Intelligent Electronics Lab and the team of Professor Kim Jin-sul of the Smart Mobile / Media Computing Lab developed an “AI-based Defective Product Identification Model” through joint research, which can identify defective products that occur in the manufacturing process at the production site.

As a result of the test verification, this model achieved 99.2% accuracy, which will not only increase the production automation rate, but also contribute to accelerating the realization of smart factories.

The research team presented these findings at the 9th International Conference on Smart Media and Applications (SMA 2020) hosted by the Smart Media Society in September and won the 'Best Paper Silver' award.

Based on the results of this study, the research team plans to continue research that can improve algorithms for improving the accuracy of deep learning models and integrate them with PLC modules at the production site.

The research team is currently conducting the Local SW Service Commercialization Support for AI-based Production Site PLC Module and Control Program Development Project supervised by the Ministry of Science, and ICT, and the National IT Industry Promotion Agency. It is cooperating with Gwangju Metropolitan City and the Gwangju Information and Culture Industry Promotion Agency to work further on the task.