Object identification and inventory management

• US 9,646,206 B2
• Filed: 08/31/2015
• Issued: 05/09/2017
• Est. Priority Date: 09/15/2011
• Status: Active Grant

• Current Assignee: Alitheon, Inc.
• Original Assignee: Alitheon, Inc.
• Inventors: Ross, David Justin, Elmenhurst, Brian J.
• Primary Examiner(s): Mistry, Oneal R
• Application Number:
  US14/841,359
• Publication Number:
  US 20150371087A1
• Time in Patent Office: 617 Days
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06V 20/80 Recognising image objects characterised by unique random patterns

Abstract

A method/apparatus for identifying an object based on a pattern of structural features located in a region of interest wherein the pattern of features comprises at least one fingerprint feature. The region may be recognized and used to identify the object. A first feature vector may be extracted from a first image of the pattern and may be mapped to an object identifier. To authenticate the object, a second feature vector may be extracted from a second image taken of the same region later in time than the first image. The two feature vectors may be compared and differences between one or more feature vector values determined. A match correlation value may be calculated based on the difference(s). The differences may be dampened if associated with expected wear and tear. Thus the impact on the match correlation value of such differences may be reduced. The differences may be enhanced if associated with changes that are not explainable as wear and tear. Thus the impact on the match correlation value of such differences may be increased.

Claims

1. A machine-implemented method comprising:

selecting a region of interest on an object;

capturing a first image of the selected region of interest so as to form first image data, wherein the first image has sufficient resolution to show at least one fingerprint feature that is innate to the object itself, a result of a manufacturing process, a result of external processes, or of any other random or pseudo random process;

processing the first image data to form a first feature vector expressing the fingerprint feature, wherein the first feature vector comprises a first set of numeric values based on the first image data, and the feature vector does not express a unique serial number of the object;

storing the first feature vector in a database record;

after said storing, performing a process with respect to an article, the process comprising;

selecting a region on the article, the selected region of the article corresponding to the region of interest of the object;

capturing a second image of the selected region of the article, so as to form second image data;

processing the second image data to form a second feature vector, wherein the second feature vector comprises a second set of numeric values based on the second image data, and the second feature vector does not express a serial number or other identifier of the article;

searching the database using the second feature vector for search criteria to find potentially matching records;

for each potentially matching record in the database, accessing a feature vector stored in the corresponding record;

comparing the stored feature vector to the second feature vector, by comparing the first set of numeric values to the second set of numeric values to determine difference values;

identifying a match between the object and the article in the case that the difference values are below a predetermined threshold; and

generating a result message indicating that the article matches a record previously stored in the database.

2. The method of claim 1 further comprising calculating the difference values asymmetrically, including:

determining whether the first feature vector indicates a feature that is absent from the second feature vector; and

reducing the difference values based on a result of the determination.

3. The method of claim 1 further comprising calculating the difference values asymmetrically, including:

determining whether the second feature vector indicates a feature that is absent from the first feature vector; and

increasing the difference values based on a result of the determination.

4. The method of claim 1 further comprising reporting a match in the case that either the difference value is below a threshold or a next-lowest difference value is sufficiently larger than a lowest difference value.

5. The method of claim 1 further comprising dampening the match correlation value by eliminating ones of the difference values that are below a threshold value.

6. The method of claim 1 further comprising dampening the match correlation value by reducing ones of the difference values that are below a threshold value.

7. The method of claim 1 wherein increasing at least one of the difference values further comprises increasing ones of the difference values that are above a threshold value.

8. The method of claim 1 wherein the fingerprint feature comprises at least one of:

cracks or other non-uniformities in surface material, a crystalline pattern, a pattern of fibers, a bleed pattern, a pattern of fabrication irregularities.

9. The method of claim 1 wherein extracting the first feature vector includes selecting an average grayscale level in each sub-region of the first image as the corresponding numeric value for that sub-region, and selecting an average grayscale value in each sub-region of the second image as the corresponding numeric value for that sub-region.

10. The method of claim 9 wherein the comparing step includes calculating a sum of the squares of the difference values, and then comparing the sums of squares.

11. The method of claim 1 wherein modifying includes, only in the case that a square of an individual difference in corresponding numeric values is below a predetermined threshold value, reducing a corresponding one of the difference values before calculating the sum of the squares of the difference values, thereby dampening the difference.

12. The method of claim 1 including storing the location of the region of interest in the database and maintaining the location as confidential.

13. The method of claim 1 wherein the object is one of a weapon, a document, a coin, a gemstone, an animal, and an art work.

18. The method of claim 1 wherein:

selecting the region of interest on the object includes first identifying a structure on the object, and then selecting the region of interest based on a predetermined offset from the structure.

19. The method of claim 1 wherein modifying includes, only in the case that a square of an individual difference in corresponding numeric values is above a predetermined threshold value, reducing a corresponding one of the difference values before calculating the sum of the squares of the difference values, thereby dampening the difference.

14. A machine-implemented method comprising:

capturing a first digital image of a first selected region of an object, wherein the digital image has sufficient resolution to show an area in the first selected region comprising a first pattern of features;

processing the first digital image to generate a first feature vector comprising data corresponding to the first pattern of features of the object;

storing the first feature vector and the first digital image in a database in association with an object identifier; and

after said storing, performing a process with respect to an article, the process comprising;

selecting a second selected region of the article, the second selected region of the article corresponding to the first selected region of the object;

capturing a second digital image of the second selected region of the article, wherein the second digital image has sufficient resolution to show an area comprising a second pattern of features;

processing the second digital image to generate a second feature vector comprising data corresponding to the second pattern of features of the article;

comparing the first feature vector with the second feature vector to determine difference values;

responsive to the comparison, augmenting or modifying a difference between the first and second feature vectors, the augmenting or modifying comprising at least one of reducing at least one of said difference values or increasing at least one of said difference values;

determining a match confidence that the data corresponding to the second pattern of features matches the data corresponding to the first pattern of features responsive to the augmenting or modifying;

comparing the match confidence to a predetermined threshold; and

in the case that the match confidence exceeds the predetermined threshold, indicating that the article is the object.

15. The method of claim 14 wherein the first pattern of features is a natural product of manufacture of the object or manufacture of a material or piece incorporated into the object.

16. The method of claim 14 wherein the first pattern of features comprises metal crystal structure defects or variations, the data of the first feature vector indicating a location of at least one defect or variation in a crystalline structure of a metal region recorded in the digital image.

17. The method of claim 14 wherein the first pattern of features comprises metal machining or milling marks presumably resulting from manufacture of a material or piece incorporated into the object.