Systems and Methods for Authenticating Jewelry and/or Gemstones
A variety of techniques for authenticating jewelry and gemstones is provided. With some examples, a stored reference signature for an item is accessed, where the item comprises jewelry and/or a gemstone, where the reference signature comprises feature values for a set of features about the item, and where the feature values are derived from previously sensed physical characteristic data about the item. A test item is sensed to generate sensed physical characteristic data about the test item. This sensed physical characteristic data about the test item is processed to extract new feature values for the set of features, where the new feature values serve as a test signature of the test item. One or more processors can compare the test signature with the accessed reference signature and authenticate the test item as being the same as the item if the comparing indicates that the test signature and reference signature match.
1 . A system for applying artificial intelligence (AI) to support authentication of jewelry and/or gemstones, the system comprising:
a sensor system configured to sense a test item to generate sensed physical characteristic data about the test item; and
one or more processors configured to:
access a stored reference signature for an item, wherein the item comprises jewelry and/or a gemstone, wherein the reference signature comprises a plurality of feature values for a set of features about the item, wherein the feature values are derived from previously sensed physical characteristic data about the item;
process sensed physical characteristic data about the test item to extract a plurality of new feature values for the set of features, wherein the new feature values serve as a test signature of the test item;
compare the test signature with the accessed reference signature; and
authenticate the test item as being the same as the item if the comparison indicates that the test signature matches the reference signature.
2 . The system of claim 1 wherein the sensed physical characteristic data comprises optical characteristic data derived from one or more images of the test item, wherein the feature values are derived from one or more previous images that depict the item, and wherein the one or more processors are further configured to process one or more new images of the test item to extract the new feature values.
3 . The system of claim 2 wherein the sensor system comprises a camera system that is configured to image the test item from a plurality of angles to capture a plurality of new images of the test item that are processed to extract the new feature values.
4 . The system of claim 2 wherein the item is marked with indicia, and wherein one or more of the previous images and one or more of the new images depict the marked item in combination with the indicia as marked thereon. The system of claim 1 wherein the reference signature uniquely identifies the item.
6 . The system of claim 1 wherein the sensed physical characteristic data comprises sensed acoustical, electromagnetic, scanning electron microscopy, and/or diffractometry characteristic data.
7 . The system of claim 1 wherein the set of features to use for the reference signature were determined based on an AI, ML, and/or computer vision system that is trained to recognize objects.
8 . The system of claim 7 wherein the AI, ML, and/or computer vision system is a trained AI system that is trained based on a plurality of known sensed physical characteristic data for jewelry and/or gemstones to learn how to distinguish between different items of jewelry and/or gemstones.
9 . The system of claim 8 wherein the trained AI system comprises a convolutional neural network (CNN).
10 . The system of claim 9 wherein the CNN comprises a deep learning CNN.
11 . The system of claim 7 wherein the features included in the set of features comprise features that are found by machine learning to correlate to distinguishing between different items of jewelry and/or gemstones from each other based on one or more defined correlation thresholds.
12 . The system of claim 11 wherein the correlated features further comprise features that exhibit sufficient variance in their feature values as between different items of jewelry and/or gemstones based on one or more variance thresholds.
13 . The system of claim 1 wherein the set of features includes a plurality of features that represent a plurality of three-dimensional (3D) characteristics of the item.
14 . The system of claim 1 wherein the one or more processors are further configured to access the stored reference signature from one or more blockchains.
15 . The system of claim 14 wherein the authentication comprises a plurality of node operators on the one or more blockchains performing authentication with respect to different portions of the accessed reference signature according to secure multi-party computation (SMPC).
16 . The system of claim 1 wherein the one or more processors are further configured to:
access stored user identification data for the item;
receive test user identification data for a seller of the test item;
compare the accessed stored user identification data with the received test user identification data to determine whether a match exists; and
condition a transaction of the test item on (1) the authentication resulting in a determination that the test signature matches the reference signature and (2) the user identification data comparison resulting in a determination that the accessed stored user identification data matches the received test user identification data.
17 . The system of claim 16 wherein the stored user identification data comprises biometric data.
18 . The system of claim 17 wherein the biometric data is based on facial recognition, a fingerprint, and/or a retina scan.
19 . The system of claim 16 wherein the one or more processors are further configured to access the stored user identification data for the item from one or more blockchains.
20 . The system of claim 1 wherein the one or more processors are further configured to access the stored reference signature based on a link in a digital asset.
21 . The system of claim 20 wherein the digital asset comprises a non-fungible token (NFT).
22 . The system of claim 1 wherein the one or more processors comprises a plurality of processors, wherein the processors include different processors that perform the access and process operations.
23 . A method for applying artificial intelligence (AI) to support authentication of jewelry and/or gemstones, the method comprising:
accessing a stored reference signature for an item, wherein the item comprises jewelry and/or a gemstone, wherein the reference signature comprises a plurality of feature values for a set of features about the item, wherein the feature values are derived from previously sensed physical characteristic data about the item;
sensing a test item to generate sensed physical characteristic data about the test item;
processing sensed physical characteristic data about the test item to extract a plurality of new feature values for the set of features, wherein the new feature values serve as a test signature of the test item;
comparing the test signature with the accessed reference signature; and
authenticating the test item as being the same as the item if the comparing indicates that the test signature matches the reference signature; and
wherein the accessing, processing, comparing, and authenticating steps are performed by one or more processors.
24 . A computer program product for supporting authentication of jewelry and/or gemstones, the computer program product comprising:
machine-readable code for a mobile application that is resident on a non-transitory computer-readable storage medium, wherein the code, in response to execution by one or more processors of a mobile computing device, is configured to cause the one or more processors to:
read a link in a digital asset to request access to a stored reference signature for an item, wherein the item comprises jewelry and/or a gemstone, wherein the reference signature comprises a plurality of feature values for a set of features about the item, wherein the feature values are derived from previously sensed physical characteristic data about the item;
request an authentication of a test item based on the accessed reference signature and a test signature for the test item, wherein the test signature comprises a plurality of new feature values for the set of features, wherein the new feature values are derived from sensed physical characteristic data about the test item; and
provide data for display to a user that is indicative of whether the requested authentication resulted in the test item being authenticated as the item based on the test signature and the reference signature.
25 . The computer program product of claim 24 wherein the sensed physical characteristic data comprises optical characteristic data derived from one or more images of the test item, wherein the feature values are derived from one or more previous images that depict the item, and wherein the new feature values are derived from one or more images that depict the test item.
26 . The computer program product of claim 25 wherein the code, in response to execution by one or more processors of the mobile computing device, is further configured to cause the one or more processors to instruct a camera of the mobile computing device to capture one or more images of the test item from which the test signature is derived.
27 . The computer program product of claim 24 wherein the set of features to use for the reference signature were determined based on an AI, ML, and/or computer vision system that is trained to recognize objects.
28 . The computer program product of claim 24 wherein the digital asset comprises a non-fungible token (NFT), and wherein the code, in response to execution by one or more processors of the mobile computing device, is further configured to cause the one or more processors to read a link in the NFT to request access to the stored reference signature from one or more blockchains.
29 . The computer program product of claim 24 wherein the authentication is performed by one or more processors that are remote from the mobile computing device.
30 . The computer program product of claim 24 wherein the code, in response to execution by one or more processors of the mobile computing device, is further configured to cause the one or more processors to:
read the link in the digital asset to request access to stored user identification data for an owner of the item;
receive test user identification data from a seller of the test item; and
request an authentication of the seller based on the accessed user identification data for the owner and the test user identification data; and
wherein the provided data for display is indicative of whether the requested authentications resulted in (1) the test item being authenticated as the item based on the test signature and the reference signature and (2) the seller being authenticated as the owner based on the user identification data for the owner and the test user identification data.