Using Artificial Intelligence to Inspect Batteries with 3D X-ray Microscopy and Computed Tomography

This presentation shows how to use artificial intelligence technologies to inspect batteries with 3D X-ray microscopy (XRM) and computed tomography (CT). More specifically, it illustrates how deep learning-based algorithms for CT reconstruction can be integrated into 3D X-ray inspection workflows for batteries. In addition, artificial intelligence (AI) provides a wide range of tools for automated defect recognition (ADR) of complex components that are difficult to analyze using traditional measurement methods. Machine learning (ML) platforms can train ADR models to assess battery overhang and inclusions.

• ZEISS DeepRecon, allows for 3D XRM workflows to be applied much more economically by reducing the time required for data acquisition

• ML for defect detection provides a wide range of tools that can increase accuracy of detection and allow for ADR to work on lower quality (faster) CT images

• Artificial intelligence-based X-ray inspection technologies will have a major impact on testing and failure analysis of batteries where non-destructive imaging is often required