Identification method and apparatus

• US 20040234108A1
• Filed: 05/22/2003
• Published: 11/25/2004
• Est. Priority Date: 05/22/2003
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Li, Dongge, Gandhi, Bhavan
• Granted Patent: US 7,272,246 B2
• Time in Patent Office: Days
• US Class Current: 382/116
• CPC Class Codes: G06V 10/768 using context analysis, e.g. recognition aided by known co-occurring patterns
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

A processor (10) utilizes information regarding one or more physical dimensions of an individual (14) to better inform a personal identification process. In one embodiment, the measured physical dimensions are utilized to influence the conduct of a face recognition process. In one embodiment, a Bayesian Belief Network can be utilized to facilitate such processes.

Claims

1. A method comprising:

detecting a human face;

detecting at least one other physical dimension that corresponds to a body of a person having the human face;

conducting a personal identification process as a function, at least in part, of the human face and the at least one other physical dimension.

2. The method of claim 1 wherein detecting at least one other physical dimension that corresponds to a body of a person having the human face includes determining at least a linear measurement that corresponds to the body.

3. The method of claim 2 wherein determining at least a linear measurement that corresponds to the body includes determining at least an approximate height of the body.

4. The method of claim 2 wherein determining at least a linear measurement that corresponds to the body includes:

capturing an image of the body; and

processing the image to at least estimate the linear measurement.

5. The method of claim 4 wherein capturing an image of the body includes capturing at least one image of the body and wherein processing the image includes processing at least one image of the at least one image of the body to at least estimate the linear measurement.

6. The method of claim 5 wherein capturing an image of the body includes:

capturing an image of the body at a moment that is substantially coincident with detection of the person'"'"'s weight; and

capturing an image of the body using only a single image capture device.

7. The method of claim 1 wherein detecting at least one other physical dimension that corresponds to a body of a person having the human face includes determining at least a weight measurement that corresponds to the body.

8. The method of claim 7 wherein determining at least a weight measurement that corresponds to the body includes determining at least an approximate weight that corresponds to the entire body.

9. The method of claim 8 wherein determining at least an approximate weight that corresponds to the entire body includes using a weight sensor to determine the weight that corresponds to the entire body.

10. The method of claim 7 wherein detecting a human face includes not detecting a human face until determining the weight measurement.

11. The method of claim 10 wherein determining at least a weight measurement that corresponds to the body includes determining that the weight measurement at least exceeds a predetermined trigger threshold.

12. The method of claim 1 wherein detecting at least one other physical dimension that corresponds to a body of a person having the human face includes detecting at least two physical dimensions that correspond to a body of a person having the human face.

13. The method of claim 12 wherein detecting at least two physical dimensions that correspond to a body of a person having the human face includes detecting at least an approximate height of the body and an approximate weight of the body.

14. The method of claim 1 wherein conducting a personal identification process as a function, at least in part, of the human face and the at least one other physical dimension includes determining not to conduct automatic face recognition of the human face as a function, at least in part, of the at least one other physical dimension.

15. The method of claim 1 wherein conducting a personal identification process as a function, at least in part, of the human face and the at least one other physical dimension includes determining to conduct automatic face recognition of the human face as a function, at least in part, of the at least one other physical dimension.

16. The method of claim 1 wherein conducting a personal identification process as a function, at least in part, of the human face and the at least one other physical dimension includes determining when the human face likely comprises a particular predetermined human face as a function, at least in part, of the at least one other physical dimension.

17. The method of claim 1 wherein conducting a personal identification process includes using a Bayesian Belief Network.

18. The method of claim 17 wherein using a Bayesian Belief Network includes determining a probability of likelihood that:

the human face is that of a predetermined particular individual;

the at least one other physical dimension is that of the predetermined particular individual.

19. The method of claim 1 wherein detecting a human face includes using an image capture device to capture an image that includes the human face.

20. The method of claim 19 wherein using an image capture device includes responding to an opportunity detection monitor.

21. The method of claim 20 wherein the opportunity detection monitor includes at least one of detecting a weight and detecting motion.

22. The method of claim 20 wherein responding to an opportunity detection monitor includes detecting an opportunity in any of a plurality of areas where a candidate person may be present

23. A method for automatically recognizing a human face, comprising:

automatically detecting a human face;

automatically detecting at least one other physical dimension that corresponds to a body of a person having the human face;

automatically conducting face recognition of the human face as a function, at least in part, of the at least one other physical dimension.

24. The method of claim 23 wherein automatically detecting a human face includes automatically detecting a human face in an image of a person.

25. The method of claim 23 wherein automatically detecting at least one other physical dimension that corresponds to a body of a person having the human face includes automatically detecting at least one of an external body linear measurement and a weight measurement.

26. The method of claim 25 wherein the external body linear measurement comprises height of the person.

27. The method of claim 25 wherein the weight measurement comprises a weight measurement of the person'"'"'s body.

28. The method of claim 23 wherein automatically detecting at least one other physical dimension that corresponds to a body of a person having the human face includes automatically detecting at least two other physical dimensions that correspond to the body of the person having the human face.

29. The method of claim 28 wherein the at least two other physical dimensions comprise height of the person and weight of the person.

30. The method of claim 23 wherein automatically conducting face recognition of the human face as a function, at least in part, of the at least one other physical dimension includes allocating face recognition resources as a function, at least in part, of the at least one other physical dimension.

31. The method of claim 27 wherein automatically detecting a human face includes automatically detecting a face in response to detecting another predetermined event.

32. The method of claim 31 wherein detecting another predetermined event includes at least one of detecting

weight;

and

motion within a predetermined area.

33. An apparatus comprising:

a first physical dimension measurement device;

a second physical dimension measurement device;

an automatic face similarity measurement unit that is responsive to the first physical dimension measurement device and the second physical dimension measurement device;

a Bayesian Belief Network that is operably coupled to the first and second physical dimension measurement device and the automatic face similarity measurement unit.

34. The apparatus of claim 33 and further comprising resource allocation means for automatically allocating face similarity measurement resources as a function, at least in part, of body height and body weight for a given individual.