Automatic classification of the degree of maturation for iron ore pellets
An automatic method or the classification of the Degrees of Maturation (DM) for fired iron ore pellets, independent of human intervention, based on the digital acquisition of microscopy images of polished cross sections, and their processing, analysis and classification by a suitable software procedure.
1. An automatic method for classification of a Degree of Maturation (DM) for a fired iron ore pellet, comprising:
digital acquisition of a microscopy image of a polished cross section of the fired iron ore pellet;
processing the image to generate a revised image;
analyzing the revised image to generate one or more image measurement parameters; and
classifying the image via an image classification component, including executing one or more image classification functions for:
comparing the one or more image measurement parameters for the revised image to a set of one or more corresponding image classification parameter rules corresponding to each of one or more DM classifications, wherein each of the one or more DM classifications represents a change in a microstructural characteristic of the fired iron ore pellet during thermal treatment; and
assigning the image a DM classification from the one or more DM classifications based on the one or more image measurement parameters for the revised image meeting the set of one or more corresponding image classification parameter rules corresponding to the DM classification.
2. The method of claim 1 , wherein the processing of the image includes one or more of background correction, edge enhancement, segmentation, and morphological post-processing.
3. The method of claim 1 , wherein the analyzing of the image comprises generating one or more types of measurement parameters selected from the group consisting of field parameters, size parameters, texture parameters, and shape parameters.
4. The method of claim 3 , wherein the field parameters comprise particle and pore count, particle and pore area fraction, and particle mean lineal intercept.
5. The method of claim 3 , wherein the size parameters comprise particle and pore area, perimeter, calipers, and equivalent circular diameter.
6. The method of claim 3 , wherein the texture parameters comprise a mean intensity of pixel intensities in a field image, a standard deviation of pixel intensities in a field image, and Haralick parameters measured from the pixel intensities in the field image.
7. The method of claim 3 , wherein the shape parameters comprise particle aspect ratio, convexity based on convex area and perimeter, circular shape factors based on area, perimeter and maximum caliper, branching and modification ratios based on inscribed circle diameter and minimum or maximum calipers.
8. The method of claim 1 , wherein the DM is represented by a class selected from the group consisting of A, B, C, and D and intermediate types AB, BC, and CD.
9. The method of claim 8 , wherein the DM increases from A to D.
10. The method of claim 1 ,
wherein the digital acquisition of the microscopy image is via a digital image capture device or a digital image acquisition component;
wherein the processing of the image is via an image processing component; and
wherein the analyzing of the image is via an image analysis component.
11. The method of claim 1 , wherein the method is reproducible.
12. The method of claim 1 , wherein the method is independent of an operator.
13. The method of claim 1 , wherein the method employs an automated microscopy system.
14. The method of claim 1 , wherein the set of one or more corresponding image classification parameter rules include values or ranges for one or more representative image measurement parameters corresponding to each of the one or more DM classifications.
15. The method of claim 1 , wherein the measurement parameters comprise at least one of particle and pore count, particle and pore area fraction, particle mean lineal intercept, particle and pore area, perimeter, calipers, equivalent circular diameter, a mean intensity of pixel intensities in a field image, a standard deviation of pixel intensities in a field image, Haralick parameters measured from the pixel intensities in the field image, particle aspect ratio, convexity based on convex area and perimeter, circular shape factors based on area, perimeter and maximum caliper, or branching and modification ratios based on inscribed circle diameter or minimum or maximum calipers.
16. The method of claim 8 , wherein the Degrees of Maturation increase from A to D, considering class A as the lowest Degree of Maturation and class D as the highest Degree of Maturation.