About the SurfConInspect Steel Surface Inspection Data Repository

The SurfConInspect Project aims to enable zero-defect manufacturing for flat steel production through an early detection of surface defects and a fast and adequate control action once a defect appears. The SurfConInspect (SCI) platform that is being implemented, is able to evaluate the surface quality images from advanced 3D and spectral band-specific detectors of a given steel production line, and provide in-coil control actions for the operators - with the help of Augmented Reality (AR) devices - as well as directly for the process control systems.

The Open Steel Surface Inspection Data Repository has been established to foster transparency, innovation, and efficiency within the steel industry and beyond. By making high-quality data openly available, this initiative supports collaboration, knowledge sharing, and the advancement of intelligent solutions for industrial applications.

Open data plays a key role in accelerating progress — it empowers researchers, engineers, and industry stakeholders to make informed decisions, explore new methods and architectures, and co-develop technologies that improve quality and productivity across manufacturing domains.

The ASIS image dataset provides real-world examples of steel surface conditions for research and development purposes. Sharing these images enables the exploration and comparison of different approaches to automated surface inspection, including:

• Development and benchmarking of machine learning and AI models.
• Creation of pre-trained classifiers for defect detection and surface quality assessment.
• Expansion of high-quality training datasets through collaborative data contributions.
• Support for academic and industrial research in computer vision, edge AI, and quality control.

For more information about the project: SurfConInspect