Stereo vision measurement system and method
A method for automatic, stereo vision based, in-line solder ball height and substrate coplanarity inspection includes providing an imaging setup together with an imaging processor for reliable, in-line solder ball height measurement. The imaging set up includes a pair of cameras mounted at two opposing angles with ring lighting around each camera lens, which allows the capture of two images of an electronics package in parallel. The lighting generates features on solder balls located on a substrate of the electronics package, which are then used to determine height. Specifically, points with the same intensity on each solder ball surface are grouped, which allows for the formation of curves, also known as iso-contours, which are then matched between the two views. An optimized triangulation is then performed to determine the height of each one of the solder balls.
1. A method of operating an imaging processor in an image processing system comprising:
obtaining at least one stereo image from an imaging setup in the image processing system, wherein the at least one stereo image includes a left view and a right view of an electronics package including a substrate and a plurality of solder balls on a surface of the substrate such that each one of the left view and the right view are obtained from separate cameras located at opposite angles with respect to the electronics package;
for each one of the left view and the right view, grouping a number of points on a surface of each one of the plurality of solder balls together based on the intensity of light reflected from the particular point and connecting the grouped points together in order to form a plurality of iso-contours;
visually segmenting the plurality of solder balls from the substrate such that a solder ball mask is generated, which includes a plurality of solder ball mask boundaries that delineate a boundary of each one of the plurality of solder balls with respect to the surface of the substrate;
calculating a centroid of each one of the plurality of solder ball mask boundaries; and
matching each one of the plurality of iso-contours in the right view to a corresponding iso-contour in the left view to form a plurality of matched iso-contours, wherein the matching comprises:
organizing nested iso-contours from each one of the right view and the left view into respective inclusion tree structures;
removing outliers from each one of the inclusion tree structures; and
matching branches of the inclusion tree structures to one another based on a relationship between nodes in the inclusion tree structures.
2. An imaging processor for use in an image processing system, the imaging processor comprising:
a processor;
communications circuitry;
a memory storing instructions which, when executed by the processor, cause the imaging processor to:
obtain at least one stereo image from an imaging setup in the image processing system, wherein the at least one stereo image includes a left view and a right view of an electronics package including a substrate and a plurality of solder balls on a surface of the substrate such that each one of the left view and the right view are obtained from separate cameras located at opposite angles with respect to the electronics package;
for each one of the left view and the right view, group a number of points on a surface of each one of the plurality of solder balls together based on the intensity of light reflected from the particular point and connecting the grouped points together in order to form a plurality of iso-contours;
visually segment the plurality of solder balls from the substrate such that a solder ball mask is generated, which includes a plurality of solder ball mask boundaries that delineate a boundary of each one of the plurality of solder balls with respect to the surface of the substrate;
calculate a centroid of each one of the solder ball mask boundaries; and
match each one of the plurality of iso-contours in the right view to a corresponding iso-contour in the left view to form a plurality of matched iso-contours, wherein the matching comprises:
organizing nested iso-contours from each one of the right view and the left view into respective inclusion tree structures;
removing outliers from each one of the inclusion tree structures; and
matching branches from each one of the inclusion tree structures to one another based on a relationship between nodes in the inclusion tree structures.
3. The method of claim 1 wherein visually segmenting the plurality of solder balls from the substrate comprises visually segmenting the plurality of solder balls from the substrate using an adaptive thresholding method based on histogram analysis.
4. The method of claim 1 further comprising calculating a centroid of each iso-contour in at least one of the plurality of matched iso-contours.
5. The method of claim 4 wherein the at least one of the plurality of matched iso-contours is the matched iso-contour with the largest average intensity gradient magnitude value.
6. The method of claim 5 further comprising triangulating a distance between the centroid of each iso-contour in the at least one of the plurality of matched iso-contours and the centroid of a corresponding one of the solder ball mask boundaries to determine a height of a solder ball.
7. The method of claim 4 further comprising triangulating a distance between the centroid of each iso-contour in the at least one of the plurality of matched iso-contours and the centroid of a corresponding one of the solder ball mask boundaries to determine a height of a solder ball.
8. The imaging processor of claim 2 wherein visually segmenting the plurality of solder balls from the substrate comprises visually segmenting the plurality of solder balls from the substrate using an adaptive thresholding method based on histogram analysis.
9. The imaging processor of claim 2 wherein the memory further includes instructions which, when executed by the processor, cause the imaging processor to calculate a centroid of each iso-contour in at least one of the plurality of matched iso-contours.
10. The imaging processor of claim 9 wherein the at least one of the plurality of matched iso-contours is the matched iso-contour with the largest average intensity gradient magnitude value.
11. The imaging processor of claim 10 wherein the memory further includes instructions which, when executed by the processor, cause the imaging processor to triangulate a distance between the centroid of each iso-contour in the at least one of the plurality of matched iso-contours and the centroid of a corresponding one of the solder ball mask boundaries to determine a height of a solder ball.
12. The imaging processor of claim 9 wherein the memory further includes instructions which, when executed by the processor, cause the imaging processor to triangulate a distance between the centroid of each iso-contour in the at least one of the plurality of matched iso-contours and the centroid of a corresponding one of the solder ball mask boundaries to determine a height of a solder ball.