Research and Development Project titled: Research and development of an artificial intelligence system for the detection of defects in product appearance
EXP 00156124 / IDI-20230527 – CONTER CONTROL DE ENERGÍA S.A.
Co-Financing and Communication
This project has been co-financed by the European Union through the European Regional Development Fund (ERDF), under the Spain 2021–2027 Pluriregional Programme. The management of this aid is overseen by the Centre for Technological Development and Innovation (CDTI), a public entity under the Ministry of Science and Innovation. Under the slogan ‘Europa se siente’ this co-financing contributes to boosting research, technological development, and innovation in Spain.
CONTER CONTROL DE ENERGÍA S.A. has carried out an R&D project aimed at developing a technology based on Artificial Intelligence, Deep Learning, and computer vision capable of detecting aesthetic defects (dents, scratches, scrapes, etc.) in vehicle parts, both interior and exterior, made from materials such as plastics, metals, fabrics, leathers, or woods. This solution is designed to significantly enhance current quality control systems in the automotive sector, which predominantly rely on human inspection or conventional technologies.
The project was executed in the province of Barcelona over 24 months, from January 1, 2023, to December 31, 2024, with a total budget of €666,384, of which €566,426.40 were co-funded by the European Union through the European Regional Development Fund (ERDF), under the Spain 2021–2027 Pluriregional Programme, within the framework of partially reimbursable aid for individual R&D projects managed by the Centre for Technological Development and Innovation (CDTI), a public entity under the Ministry of Science and Innovation. This initiative contributes to EU Policy Objective 1: A more competitive and smarter Europe.
Project Objectives:
- Develop an AI-based evaluation system capable of identifying aesthetic defects in industrial production.
- Design a flexible, scalable software platform communicable via API with other client environment solutions.
- Implement an efficient workflow for generating and training AI models with minimal resource consumption.
- Validate a functional prototype in a laboratory environment (TRL4), preparing it for future industrialization.
Achieved Results:
- Development of a stable and functional software platform with an intuitive interface, proprietary database, and tools for statistical analysis and maintenance.
- Integration of an AI evaluation engine based on convolutional networks, supplemented by a decision phase based on parametrized filter logic, enhancing system accuracy and traceability.
- Attainment of advanced technical knowledge in defect detection and model training, opening new business opportunities for CONTER in the field of quality control automation.