Martensitic Microstructures. Latest Clemex grain size development using machine learning capabilities. Requires Vision Lite, Core or Unlimited.
The Grain Size Application Package for Clemex Vision leverages machine learning capabilities, including a special instruction for martensitic grains. The new automatic algorithm provides a fast, reliable, and accurate way of optically identifying challenging martensitic microstructures.
Martensitic microstructures are difficult to analyze optically because of the materials’ resistance to etching techniques. Grain structures are barely visible and typically a lab will rely on the human eye to distinguish and classify them.
The gray thresholding technique used in image analysis software is not appropriate for this kind of image, as it is based on the gray level distribution of pixels. It works well with clearly defined phases that form distinct peaks on the thresholding graph.
Clemex Vision PE now has a way of processing images of various martensitic structures which does not depend on gray level thresholding. This breakthrough was accomplished after months of work on hundreds of images from industry partners.
The Grain Size Application Package for Clemex Vision leverages machine learning capabilities, including a special instruction for martensitic grains. The new automatic algorithm provides a fast, reliable, and accurate way of optically identifying challenging martensitic microstructures.
Martensitic microstructures are difficult to analyze optically because of the materials’ resistance to etching techniques. Grain structures are barely visible and typically a lab will rely on the human eye to distinguish and classify them.
The gray thresholding technique used in image analysis software is not appropriate for this kind of image, as it is based on the gray level distribution of pixels. It works well with clearly defined phases that form distinct peaks on the thresholding graph.
Clemex Vision PE now has a way of processing images of various martensitic structures which does not depend on gray level thresholding. This breakthrough was accomplished after months of work on hundreds of images from industry partners.
The Grain Size Application Package for Clemex Vision leverages machine learning capabilities, including a special instruction for martensitic grains. The new automatic algorithm provides a fast, reliable, and accurate way of optically identifying challenging martensitic microstructures.
Martensitic microstructures are difficult to analyze optically because of the materials’ resistance to etching techniques. Grain structures are barely visible and typically a lab will rely on the human eye to distinguish and classify them.
The gray thresholding technique used in image analysis software is not appropriate for this kind of image, as it is based on the gray level distribution of pixels. It works well with clearly defined phases that form distinct peaks on the thresholding graph.
Clemex Vision PE now has a way of processing images of various martensitic structures which does not depend on gray level thresholding. This breakthrough was accomplished after months of work on hundreds of images from industry partners.
ADVANTAGES
MARTENSITIC GRAIN SIZE ALGORITHM
SIMPLE WORKFLOW
The no-threshold method is a quick and easy way of detecting grain boundaries automatically
ROBUST
Repeatable detection of grains without visible boundaries even with light intensity variation
ACCURATE
Successfully tested on hundreds of images in collaboration with a leading industry partner
FAST ANALYSIS
Once the parameters are set, the algorithm is deployed on each field of the sample at high speed
REPRODUCIBILITY
The method minimizes inter-operator variability and increases reproducibility