AMBIT Living Spaces Cluster

CICLOP 2.0

Optimización de un sistema de visión artificial para el aseguramiento de calidad en procesos de impresión digital de revestimientos de gran formato a través de algoritmos de aprendizaje profundo.

Resum

Within the framework of quality perception in the market as a differentiating factor for customers, the CICLOP 2.0 project is integrated. The evolution of artificial vision systems, greatly enhanced by the image processing capabilities of deep learning algorithms, enables the development of quality control applications in which the system itself can learn and make decisions based on captured images. In an industry with limited adoption of advanced data processing technologies, the CICLOP 2.0 project aims to develop a quality control system by combining Artificial Vision with Deep Learning and facilitate its implementation in the digital printing sector.

While the initial phase of the CICLOP project aimed to capture large-format images using artificial vision for the detection of printing defects, the second phase seeks to investigate the categorization of defects through the integration of deep learning technology. This need for defect analysis to align quality criteria with customer perception forms the framework of the CICLOP 2.0 project. With the clear goal of reducing the number of incidents in the market, the project will explore how deep learning algorithms can enhance inspection reliability throughout its duration.

Transferable results to companies

The results of the project will be applied to the digitization and process improvement in the value chain of digital printing systems:

Manufacturers and distributors of decorative panels will have a technological integration based on quality inspection through deep learning algorithms.
End users, builders, and interior designers will have reliable panels and coverings produced under quality control systems that are not based on sampling but on complete production assurance.
Integrators and developers of Deep Learning technology will be able to make the necessary technological leap for the integration of image processing algorithms into applications for decorative products with varying designs and large formats.
Manufacturers of digital printing equipment will have a technology validation system. In addition to benefiting from technology that allows them to assess the condition of their equipment, the implementation of artificial vision and Deep Learning systems should facilitate interventions and equipment adjustments for the optimization of printing processes.