Quality control of carbon parts based on machine vision

Quality control of carbon parts based on machine vision

The main objective is to increase the agility level of Composite Industries production line, by automating its quality inspection process of carbon fibre components by employing a machine vision system.

The solution

To automate the quality inspection process of CFRP parts, ATLANTES proposes a solution based on a machine vision system consisting of both a traditional RGB camera and a near-infrared (NIR) hyperspectral one, to cover a wider range of possible defects and detect them better. The two technologies (RGB and NIR) are considered separately, having developed acquisition and detection codes for each of them. From the RGB point of view, the neural networks implemented by the TRINITY-originated module “Object Classification” is exploited to detect pitting, weave, and surface matte defects at a high resolution. Regarding NIR inspection, the spectroscopy of the pixels of the CFRP part is statistically analysed to detect the defective areas.


Small manufacturers of prepreg carbon-fibre-reinforced polymers (CFRP) parts have the strong need to stay agile in their offering and to always seek technologies to improve productivity. In particular, the quality inspection process of CFRP parts, which is commonly based on human visual observation, is time-consuming, expensive and fallible. Moreover, the CFRP components market is characterised by many small batches, consequently the production process changes frequently and needs to be flexible and easily reprogrammable. To address these needs, ATLANTES aims at automating the quality inspection process of CFRP parts.


Thanks to ATLANTES, the quality inspection process of CFRP parts is more agile, thanks to the introduction of; automation and being more productive and less expensive, less human labour is wasted in the visual quality inspection. In addition, the quality of the inspection is enhanced compared to human visual inspection thanks to the detection of smaller defects with better accuracy.

Facts and figures

  • Technology Area:

    Robot Programming

  • End User:

    Composite Industries who is an SME manufacturer of CFRP components

  • Start Date - End Date:

    01/11/2021 - 31/08/2022

  • Duration:

    10 months

  • FSTP Funding:

    199 938,00 €

  • TRL Level at Start:


  • TRL Level at End:


  • Number of early adopters raised:


  • Project results
    The machine vision system is very useful for improving the quality inspection process of our components, as it allows to detect defective areas at a higher accuracy than the traditional visual process.

    Lidia Iulia Pop, Quality Manager at Composite Industries

    back to success stories
    Subscribe to newsletter

    Please subscribe to our mailing list to be the first one informed about the new open calls, events and other hot news!

      Please check if the email you entered is correct
      All done!

      Your have successfully subscribed to Trinity newsletter.