AI Application for Digitization and Digitalization
MICROSCOPY SOLUTIONS FOR NATURAL RESOURCES

AI Application for Digitization and Digitalization

Streamlined workflows. Realized.

Continued algorithmic innovation, coming from the distinctive AI infrastructure at ZEISS, ensures that ZEISS microscope users benefit from advancements in technology and enhanced data evaluation.

Applying AI for automated image segmentation or registration, for enhanced image quality and resolution, and for quantitative analyses of the rich underlying data, helps geoscientists by illuminating the relevant information needed for better data-driven decision making.

  • Enhance image and data filtering and segmentation
  • Automate the quantitative measurement of your samples
  • Streamline the path to results
  • Analyze more data, more efficiently, with confidence

ZEISS Integrates Machine Learning in Multi-Scale Microscopy for Enhanced Natural Resources Workflows

Natural rock samples can be very complex requiring observations at multiple length scales within a single project. This brings a number of challenges across all the microscopy techniques as we balance the requirements of our task with sample size and resolution. Implicit within this challenge are the problems of file size, and data processing time. ZEISS incorporates machine learning algorithms within software workflows for light, electron and X-ray microscopy. Here we describe how each of those software tools are incorporated into Natural Resources workflows for rock, mineral, and related samples across the Earth Sciences in Industry and Academia.

Advanced Reconstruction Toolbox (ART)

Advanced Reconstruction Toolbox (ART)

The ZEISS Advanced Reconstruction Toolbox is a suite of capabilities for ZEISS micro computed tomography (microCT) and X-ray Microscopy (XRM) instruments. DeepRecon Pro increases the contrast-to-noise ratio of 3D reconstructions enabling higher clarity images and higher sample throughput, while DeepScout provides the capability for upscaling the resolution of images.

Automated Quantitative Mineralogy

Automated Quantitative Mineralogy

ZEISS Mineralogic 3D is a revolutionary, quantitative approach to microCT that enables consistent, predictable absorption contrast values for scanned materials in 3D. This technology means that within geological samples mineral phases can be identified directly from absorption contrast data whilst leaving the sample intact. ZEISS DeepRecon Pro is an integral part of the ZEISS Mineralogic 3D workflow where the increase in contrast-to-noise ratio enhances the ability to distinguish similar mineral phases, a key challenge for microCT. Powered by deep learning image enhancement Mineralogic 3D allows for streamlined mineral identification and segmentation of your entire sample with minimal user input across large projects or process mineralogy.

ZEN AI Toolkit and Automated Image Segmentation

Rapidly and intuitively label regions of interest, including applying mineral names, identifying porosity, and many other key workflows.

ZEN AI Toolkit and Automated Image Segmentation

Bituminous coal, brightfield light microscopy with Axio imager Z2.m

Bituminous coal, brightfield light microscopy with Axio imager Z2.m

ZEN AI Toolkit and Automated Image Segmentation

Geoscientists of all kinds generate quantitative data across a wide range of samples and imaging types. The first step in generating quantitative information for mineralogy and textural analysis is image segmentation. ZEISS ZEN Intellesis allows the user to rapidly and intuitively label regions of interest, including applying mineral names, identifying porosity, and many other key workflows. Once data have been segmented the ZEISS ZEN package can generate automated outputs for quantitative measurements of the sample including sizes, shapes and distributions of mineralogy and other components.
Combining ZEN Intellesis with the Petrography Analysis Toolbox for Axioscan 7 Geo allows for segmentation and orientation analysis. This powerful combination uses the power of Axioscan 7 Geo data to generate a comprehensive mineral-textural analysis of each sample.

ZEISS arivis Pro and arivis Cloud

ZEISS arivis Pro and arivis Cloud

ZEISS arivis is a powerful and comprehensive image analysis platform that can bring and combine machine learning enhanced workflows on your workstation and in the cloud.

ZEISS arivis image analysis includes instance segmentation, a power machine learning tool that builds upon what is possible with pixel classification. Where pixel classification provides a mechanism to identify mineralogy, “instance segmentation” allows the user to identify a type of object. For geoscience workflows in industry and academia this means samples can be segmented by lithology. This is a key advancement as classification by rock type is a common workflow that can now be automated from identification to reporting with the same level of detail as mineralogy.

Automated Quantitative Mineralogy
USER STORY

From the Tiniest Particles to the Greatest Discoveries

Artificial Intelligence in Microscopy

To understand the bigger picture, Dr. Oliver Plümper focuses on very small things: the geoscientist studies minerals that are components of rocks. But what stories do these components tell – and how do they help us understand the secrets of our planet Earth? To find this out, Plümper relies on engineering and technology: microscopes – and artificial intelligence.

Downloads

    • Assessing grit-blasted metal surfaces by confocal light microscopy and scanning electron microscopy

      Assess grit-blasted metal surfaces by confocal light microscopy and scanning electron microscopy

      File size: 4 MB
    • ZEISS Mineralogic 3D

      Whole-particle Liberation Studies

      File size: 1 MB

Visit the ZEISS Download Center for available translations and further manuals.

Contact ZEISS Microscopy

Contact

Form is loading...

/ 4
Next Step:
  • Step 1
  • Step 2
  • Step 3
Contact us
Required Information
Optional Information

If you want to have more information on data processing at ZEISS please refer to our data privacy notice.