Component tracking in automated manufacturing using digital fingerprints

A method for tracking components in an automated manufacturing line, comprising:

• receiving an indication that an apparatus in a first assembled state should comprise a first component with a first digital fingerprint comprising a plurality of feature points, and a second component with a second digital fingerprint comprising a plurality of feature points that are different from those of the first fingerprint, the feature points in the first and second fingerprints comprising at least one selected from an edge, a corner, a blob, or a ridge;
• receiving video footage from a camera configured to record video of apparatuses in the first assembled state, wherein the first assembled state comprises;
• a first component of a first type;
• and
• a second component, being one of the first type or a second type;
• converting the video footage to greyscale;
• isolating, from the greyscale footage, a first frame comprising an image of an apparatus in the first assembled state;
• splitting the first frame into a second frame comprising the first component of the first type, and a third frame comprising the second component;
• applying at least one filtering algorithm to the second and third frames to generate a first and second filtered image;
• generating, using a feature detection algorithm, a first set of feature points from the first filtered image and a second set of feature points from the second filtered image, the first and second sets of feature points each comprising at least one selected from an edge, a corner, a blob, or a ridge;
• determining that the first set of feature points from the first filtered image matches the feature points comprising the first digital fingerprint;
• determining that the second set of feature points from the second filtered image matches the feature points comprising the second digital fingerprint;
• updating a component database with:
• a third digital fingerprint based on the apparatus in the first assembled state,
• a date and time when the apparatus in the first assembled state was assembled;
• and
• an indication that the apparatus in the first assembled state is associated with the first and second digital fingerprints.

A computer program comprising executable instructions stored in a non-transitory computer readable medium that when executed by a hardware processor cause the hardware processor to:

• receive an indication that an apparatus in a first assembled state should comprise a first component with a first digital fingerprint comprising a plurality of feature points, and a second component with a second digital fingerprint comprising a plurality of feature points that are different from those of the first fingerprint, the feature points in the first and second fingerprints comprising at least one selected from an edge, a corner, a blob, or a ridge;
• receive video footage from a camera configured to record video of apparatuses in the first assembled state, wherein the first assembled state comprises;
• a first component of a first type;
• and
• a second component, being one of the first type or a second type;
• convert the video footage to greyscale;
• isolate, from the greyscale footage, a first frame comprising an image of an apparatus in the first assembled state;
• split the first frame into a second frame comprising the first component of the first type, and a third frame comprising the second component;
• apply at least one filtering algorithm to the second and third frames to generate a first and second filtered image;
• generate, using a feature detection algorithm, a first set of feature points from the first filtered image and a second set of feature points from the second filtered image, the first and second sets of feature points each comprising at least one selected from an edge, a corner, a blob, or a ridge;
• determine that the first set of feature points from the first filtered image matches the feature points comprising the first digital fingerprint;
• determine that the second set of feature points from the second filtered image matches the feature points comprising the second digital fingerprint;
• update a component database with:
• a third digital fingerprint based on the apparatus in the first assembled state,
• a date and time when the apparatus in the first assembled state was assembled;
• and
• an indication that the apparatus in the first assembled state is associated with the first and second digital fingerprints.