Service & Support

ZEISS DeepRecon for X-ray Reconstruction

From the Solutions Lab for 10X throughput improvement for ZEISS Xradia X-ray microscopes for repetitive workflows

Highlights
Workflow
Components
Contact Us
Back to Solutions Lab Overview

One of the biggest challenges associated with X-ray microscopy is that of throughput combined with image quality. Acquiring high resolution images – particularly within large samples – requires long acquisition times. These long acquisition times may limit the technology’s application to industrial processes, where time per sample is critical, or when performing in situ experiments where high temporal resolution is required.

ZEISS DeepRecon allows for extreme increases in image quality and effective acquisition time through the integration of neural networks into the image acquisition and reconstruction process. By working in collaboration with customers, networks and workflows can be customized to answer specific challenges associated with repetitive samples. The use of this technology can increase throughput by up to a factor of 10, greatly improve image quality and reduce the impact of imaging artifacts that typically affect rapid acquisition workflows.

One of the biggest challenges associated with X-ray microscopy is that of throughput combined with image quality. Acquiring high resolution images – particularly within large samples – requires long acquisition times. These long acquisition times may limit the technology’s application to industrial processes, where time per sample is critical, or when performing in situ experiments where high temporal resolution is required.

ZEISS DeepRecon allows for extreme increases in image quality and effective acquisition time through the integration of neural networks into the image acquisition and reconstruction process. By working in collaboration with customers, networks and workflows can be customized to answer specific challenges associated with repetitive samples. The use of this technology can increase throughput by up to a factor of 10, greatly improve image quality and reduce the impact of imaging artifacts that typically affect rapid acquisition workflows.

Highlights

Area of Research:

  • Semiconductor
  • Geoscience
  • Manufacturing

Sample Types:

  • Electronics
  • Semiconductor packages
  • Geological samples

Related Solutions:

  • ZEISS Phase Contrast Enhancer
Deep learning-based image quality improvement with higher throughput for Xradia X-ray microscopes
Deep learning-based image quality improvement with higher throughput for Xradia X-ray microscopes

Structure of neural network-based image quality improvement, achieved by training networks to recover image data degraded through noise, limited projection numbers or sample outside of the field of view.

Workflow

The module workflow consists of a number of steps:

  1. ZEISS DeepRecon is a Special Customer Solution (SCS), customized to the specific requirements of your use-case and sample class. First, engage with your local ZEISS representative (or fill out the below form) to start the discussion about getting DeepRecon at your facility.
  2. Send data or a sample to ZEISS XRM team.
  3. This data is used to create a customized DeepRecon model, enabling superior
  4. This model is loaded into a customized installation of ZEISS Reconstructor.
  5. Acquire data under the discussed conditions (up to 10X throughput boost relative to standard reconstruction). The customized model will be a selectable option during sample reconstruction.
Progressive image quality improvement with various reconstruction techniques for Xradia X-ray microscopes
Progressive image quality improvement with various reconstruction techniques for Xradia X-ray microscopes

Impact of differing reconstruction techniques on a sandstone sample, showing progressive image quality improvement from standard reconstruction (filtered back projection) to OptiRecon 2.0 (iterative reconstruction) and DeepRecon (neural network-based reconstruction)

Required components

  • ZEISS Xradia Versa XRM
  • ZEISS DeepRecon reconstruction technology

Need More Information?

Please contact us using the following form.

Problems ​we can help you solve: ​

  • automation of your imaging process or workflow
  • advanced image analysis and corrections
  • automation of your ROI (region of interest) search
  • tailoring flexible and adaptable workflows for your configuration
  • even, small applications that turn 20 clicks into one.

Big problems, little problems. What is your problem? Use the form on this page to tell us more.​

You may want to bookmark this site and check back frequently as we are rapidly creating new solutions.

Unable to load form.