Authentication and identification of physical objects using voting of local microstructural features

A method for object identification based on a scalar score corresponding to a set of local microstructural features of an object, the method comprising:

• capturing a plurality of digital images by a device camera of one or more surfaces corresponding to the object of a plurality of objects;
• extracting the set of local microstructural features from the plurality of digital images of the one or more surfaces of object by a feature extractor of a backend module;
• storing the plurality of objects and the set of local microstructural features in a database extracted from the one or more surfaces of a corresponding object, wherein the set of local microstructural features represent the corresponding object in the database;
• receiving a query object for identification by the backend module;
• extracting a plurality of local microstructural features from a number of surfaces of the query object using the feature extractor;
• comparing the plurality of local microstructural features of the query object with the set of local microstructural features of the plurality of objects stored in the database using a voting function, wherein the voting function is a mathematical function that indicates agreements and disagreements of the plurality of local microstructural features of the query object and the set of local microstructural features of the plurality of objects;
• generating a score for the corresponding object in response to the comparing of the plurality of local microstructural features of the query object with the set of local microstructural features of the plurality of objects;
• selecting a potential match set of the plurality of objects based on the score;
• identify a status of the query object based on the potential match set, wherein the status indicates a match of the query object with at least one object of the plurality of objects stored in the database;
• and
• recognizing a state of the query object when the query object is unrepresented in the database based on similar scores obtained from the match of the query object with each object of the plurality of objects, the similar scores indicate an absence of a maximum value of the score among scores of the query object with respect to each object.

A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations, for object identification based on a scalar score corresponding to a set of local microstructural features of an object, the operations comprising:

• capturing a plurality of digital images by a device camera of one or more surfaces corresponding to the object of a plurality of objects;
• extracting the set of local microstructural features from the plurality of digital images of the one or more surfaces of the object by a feature extractor of a backend module;
• storing in a database, the plurality of objects and the set of local microstructural features extracted from the one or more surfaces of a corresponding object, wherein the set of local microstructural features represent the corresponding object in the database;
• receiving a query object for identification by the backend module;
• extracting a plurality of local microstructural features from images of a number of surfaces of the query object using the feature extractor;
• comparing the plurality of local microstructural features of the query object with the set of local microstructural features of the plurality of objects stored in the database using a voting function, wherein the voting function is a mathematical function that indicates agreements and disagreements of the plurality of local microstructural features of the query object and the set of local microstructural features of the plurality of objects;
• generating a score for the corresponding object in response to the comparing of the plurality of local microstructural features of the query object with the set of local microstructural features of the plurality of objects;
• selecting a potential match set of the plurality of objects based on the score;
• identify a status of the query object based on the potential match set, wherein the status indicates a match of the query object with at least one object of the plurality of objects stored in the database;
• and
• recognizing a state of the query object when the query object is unrepresented in the database based on similar scores obtained from the match of the query object with each object of the plurality of objects, wherein the similar scores indicate an absence of a maximum value of the score among scores of the query object with respect to each object.