DE-102025100023-A1 - SYSTEM AND METHOD FOR QUANTIFICATION OF THE SECONDARY DORDITE ARM DISTANCE

DE102025100023A1DE 102025100023 A1DE102025100023 A1DE 102025100023A1DE-102025100023-A1

Abstract

A computer-implemented method that, when executed on data processing hardware, causes the data processing hardware to perform operations. The operations include: i) detecting clusters of dendritic arms in the metallic material and generating an image frame based on the microimage representing the clusters of dendritic arms by applying a machine learning algorithm to a microimage representing a metallic material; ii) detecting individual dendritic arms within the clusters of dendritic arms by applying a watershed algorithm to the image frame; iii) applying at least one filter to the image frame; iv) identifying at least one cluster of dendritic arms in the image frame corresponding to the at least one filter; and v) generating a cluster profile for the at least one cluster of dendritic arms, which includes at least one average dendritic arm distance (DAS) value for the at least one cluster of dendritic arms.

Inventors

Meysam Akbari
Liang Wang
Qigui Wang
Cuifen Yan

Assignees

GM Global Technology Operations LLC

Dates

Publication Date: 20260513
Application Date: 20250103
Priority Date: 20241113

Claims (10)

A computer-implemented method that, when executed on data processing hardware, causes the data processing hardware to perform operations that include: by applying a machine learning algorithm to a microimage representative of a metallic material: detecting clusters of dendritic arms in the metallic material; and generating an image frame based on the microimage that is representative of the clusters of dendritic arms; detecting individual dendritic arms within the clusters of dendritic arms by applying a watershed algorithm to the image frame; applying at least one filter to the image frame; identifying at least one cluster of dendritic arms in the image frame that corresponds to the at least one filter; and generating a cluster profile for the at least one cluster of dendritic arms that includes at least one average dendritic arm distance (DAS) value for the at least one cluster of dendritic arms.
Procedure according to Claim 1 , wherein the at least one filter includes a distance filter which requires that the distances between each pair of adjacent dendrite arms of the at least one cluster of dendrite arms lie between a minimum distance and a maximum distance.
Procedure according to Claim 1 , wherein the at least one filter includes an aspect ratio filter which requires that the aspect ratio of each individual dendrite arm within the at least one cluster of dendrite arms is greater than a minimum aspect ratio.
Procedure according to Claim 1 , wherein the at least one filter comprises an angle filter which requires that the angles between each pair of adjacent dendrite arms of the at least one cluster of dendrite arms are smaller than a maximum angle.
Procedure according to Claim 1 , wherein the at least one filter includes a position filter that requires that each individual dendrite arm within the at least one cluster of dendrite arms be positioned relative to the other dendrite arms of the at least one cluster of dendrite arms between a first boundary and a second boundary.
Procedure according to Claim 1 , wherein the at least one filter includes a quantity filter that requires that the at least one cluster of dendrite arms includes at least a minimum number of individual dendrite arms.
Procedure according to Claim 1 , wherein the cluster profile further comprises at least one selected from the group consisting of (i) a number of the at least one cluster of dendritic arms, (ii) a number of individual dendritic arms within each of the at least one cluster of dendritic arms, (iii) a width of each of the at least one cluster of dendritic arms and (iv) an average angle of individual dendritic arms within each of the at least one cluster of dendritic arms.
Procedure according to Claim 1 , wherein the cluster profile further includes a display image that identifies the at least one cluster of dendrite arms.
Procedure according to Claim 1 , where at least one filter is configurable based on user input.
Procedure according to Claim 1 , where the machine learning algorithm is trained on the basis of a large number of annotated micro-images.

Description

INTRODUCTION The information contained in this section serves to present the general context of the disclosure. Works of the inventors mentioned herein, insofar as they are described in this section, as well as aspects of the description that might not otherwise be considered prior art at the time of filing, are neither expressly nor implicitly admitted as prior art against the present disclosure. The present disclosure relates generally to the quantification of the microstructural fineness of metal castings and in particular to the automated quantification of dendritic arm spacings (DAS) in dendritic microstructures of metal castings. The resulting microstructure of cast metal vehicle components (such as engine blocks, cylinder heads, transmission parts, or the like) is generally determined by the alloy composition and, in particular, by the solidification conditions, such as local cooling rates. In many alloy compositions, the materials tend to solidify dendritically, with the resulting structure containing a multitude of dendrite arms. Examples of alloy compositions that solidify dendritically include aluminum alloy 356 (A356) and aluminum alloy 319 (A319). The relative amounts, sizes, and morphology of these phases in the cast structure are highly dependent on the casting conditions as well as the alloy composition. For example, the dendrite cell size (DCS) and the DAS, sometimes referred to as secondary dendrite arm spacing (SDAS), can be used to quantify the fineness of the casting, which in turn allows for a better understanding of the material properties. Generally, cast components with a smaller DAS have higher durability and other mechanical properties. Manual DAS measurement methods can be used to determine the physical properties of aluminum castings. For example, a linear intercept method on a micrograph can be used to measure DAS. A linear intercept method involves manually selecting three or more dendrites with visible dendrite stems per field of view and manually drawing a line from an outer edge of the first dendrite stem to an inner edge of the last dendrite stem. The distance can be recorded for each dendrite, while the number of dendrite stems counted in each measurement can also be recorded. These activities can be repeated for each field of view. SUMMARY One aspect of the disclosure provides a computer-implemented method. When the computer-implemented method is executed on the data processing hardware, it causes the data processing hardware to perform operations. The operations include: i) detecting clusters of dendritic arms in the metallic material and generating an image frame based on the microimage, representing the clusters of dendritic arms, by applying a machine learning algorithm to a microimage representing a metallic material; ii) detecting individual dendritic arms within the clusters of dendritic arms by applying a watershed algorithm to the image frame; iii) applying at least one filter to the image frame; iv) identifying at least one cluster of dendritic arms in the image frame corresponding to the at least one filter; and v) generating a cluster profile for the at least one cluster of dendritic arms, which includes at least one average dendritic arm distance (DAS) value for the at least one cluster of dendritic arms. Implementations of this aspect of the disclosure may include one or more of the following optional features. In some examples, the at least one filter includes a distance filter that requires the distances between each pair of adjacent dendrite arms of the at least one cluster of dendrite arms to lie between a minimum distance and a maximum distance. In some implementations, the at least one filter includes an aspect ratio filter that requires that the aspect ratio of each individual dendrite arm within the at least one cluster of dendrite arms is greater than a minimum aspect ratio. In some configurations, the at least one filter includes an angle filter that requires that the angles between each pair of adjacent dendrite arms of the at least one cluster of dendrite arms are smaller than a maximum angle. In some examples, the at least one filter includes a positional filter that requires each individual dendrite arm within the at least one cluster of dendrite arms to be positioned relative to the other dendrite arms of the at least one cluster of dendrite arms between a first boundary and a second boundary. In some implementations, the at least one filter includes a set filter that requires the at least one cluster of dendrite arms to include at least a minimum number of individual dendrite arms. In some configurations, the cluster profile further includes at least one selected from the group consisting of (i) a number of the at least one cluster of dendrite arms, (ii) a number of individual dendrite arms within each of the at least one cluster of dendrite arms, (iii) a width of each of the at least one cluster of dendrite arms, and (iv) an average angle o