Automated defect recognition and determinations of pore cluster compliance
A process can obtain a RT image of an article from the RT imaging system, the article comprising a reference feature, identify a first portion and a second portion of the first RT image, wherein the first portion and the second portion overlap, and classify each pixel in the first portion and the second portion using a pore classification model. A pore map can then be generated and mapped onto the article based on data received from the at least one sensor. A process can identify rejected regions on the pore map, wherein the rejected regions exceed a rejection threshold based on predefined rules. These rejected regions can then be displayed.
1 . A robotic additive manufacturing inspection system configured to identify porosity defects in manufactured articles, the robotic additive manufacturing inspection system comprising:
an additive manufacturing robot, the additive manufacturing robot comprising:
a robotic arm;
a radiographic testing (RT) imaging system mounted to the robotic arm; and
at least one sensor capable of determining a position of the robotic arm;
an image processing system, the image processing system configured to:
obtain a RT image from the RT imaging system of an article created by additive manufacturing and localization information including the position of the robotic arm when the RT image was captured;
adjust brightness values of pixels within the RT image to normalize the brightness values against brightness values in a reference image;
identify a plurality of image portions within the RT image, wherein each image portion at least partially overlaps at least one other image portion;
classify each pixel in each image portion using a pore classification binary segmentation model as “pore” or “not pore” to obtain classification of the pixels in the image portions;
generate a pore map based on applying one or more porosity detection clustering filters to the classification of the pixels in the image portions, where the pore map includes identification of pores within the image portions and measurements of the pores based on a measurement calibration of the RT imaging system; and
determine whether to identify at least one area within the RT image as a rejected region based upon the pore map, when the at least one area exceeds a rejection threshold based on a set of predetermined criteria.
2 . The robotic additive manufacturing inspection system of claim 1 , where the image processing system is further configured to:
display the pore map with the rejected regions visually identified;
receive an annotation from a user, wherein the annotation is characterized by at least a location on the pore map; and
update the pore classification binary segmentation model based on the received annotation.
3 . The robotic additive manufacturing inspection system of claim 2 , wherein the annotation comprises at least one transformation selected from a list of adjustments to contrast/brightness, zoom in, circling out of blotches, circling out of defects, use of rulers to measure distances, text notations, and pore classifications.
4 . The robotic additive manufacturing inspection system of claim 1 , wherein adjusting the brightness values of pixels within the RT image to normalize the brightness values against the brightness values in a reference image comprises setting the brightness values so that a histogram of the RT image approximately matches a histogram of the reference image.
5 . The robotic additive manufacturing inspection system of claim 1 , wherein applying porosity detection clustering filters evaluates characteristics of each pore in the pore map including convexity, circularity, and a minimum size.
6 . The robotic additive manufacturing inspection system of claim 1 , wherein applying the porosity detection clustering filters provides a center point and radius of each detected pore in the pore map.
7 . The robotic additive manufacturing inspection system of claim 1 , wherein the rejection threshold based on a set of predetermined criteria includes spacing, size, and frequency between defects in the pore map.
8 . The robotic additive manufacturing inspection system of claim 1 , wherein generate a pore map further comprises:
resolving classification of a pixel as “pore” or “not pore” that differs in image portions where the pixel is included in overlapping areas by taking the identification having the highest confidence value output by the pore classification binary segmentation model.
9 . The robotic additive manufacturing inspection system of claim 1 , wherein the article comprises a reference feature and the image processing system is further configured to obtain the measurement calibration of the RT imaging system based on the reference feature.
10 . The robotic additive manufacturing inspection system of claim 1 , wherein one or more porosity detection clustering filters includes a plurality of porosity detection clustering filters, and each porosity detection clustering filter includes a different combination of values for parameters of convexity, circularity, and minimum size.