top of page
Image by Ramón Salinero


Automatic Air Bubble detection in solar panels

Machine2learn Logo.png.png
logo AI Hub East Netherlands.png

NEED for AI:

The use of AI in this process is necessary to increase efficiency and accuracy in defect detection, leading to lower manufacturing costs and increased client satisfaction.



The project focuses on using AI to detect defects in solar panel manufacturing, reducing production costs. It aligns with the topic of "AI for Manufacturing Applications and AI-on-demand Platform Contributions".



  • Utilize a product idea for digitalizing production lines, with the first scalable use case being an automatic air bubble detection algorithm

  • Provide knowledge on the solar panel industry and a use case for the visual inspection system

  • The business model would charge per instance for detecting air bubble defects and can cover the expenses of one full-time employee if applied to at least 5 production lines

  • A strong use case where AI is necessary and can contribute to the societal challenge of energy transition, making it valuable for elaboration by the AI-hub East Netherlands

Goal of this experiment is to develop a machine learning and computer vision-based system to aid quality control in the solar panel production industry by detecting air bubble defects. The solar panel market has been growing and is expected to continue to increase, and the experiment will be conducted within Exasun located in Den Haag, with the potential to be tested in other production companies. The system will use data-driven algorithms, such as deep artificial neural networks, to detect and localize air bubbles, and subsequently, use computer vision post-processing techniques to measure the severity of the defects.

bottom of page