In the fast-paced world of industrial manufacturing, companies are under constant pressure to deliver more components, faster and with greater quality. This challenge spans a broad range of sectors, including battery production, consumer goods, pharmaceuticals, automotive and aerospace industries, and electronics. A crucial factor in meeting these demands is the ability to detect and prevent defects early in the manufacturing cycle, thereby ensuring reliable, high-performance products.

X-ray imaging, including both 2D inspection and 3D computed tomography (CT), has become an essential non-destructive tool for quality assurance, process optimization, and research and development. Central to these applications is the need for effective image segmentation to detect and quantify defects. In recent years, AI-driven segmentation techniques have gained traction, offering several advantages over traditional methods. These include minimizing operator bias, enabling the interpretation of complex or noisy datasets, and facilitating fully automated analysis workflows.

This presentation explores the current landscape of AI-based image segmentation in industrial contexts, highlighting its impact through practical case studies that demonstrate measurable benefits for manufacturers and researchers alike.