Visual object identification by computational majority voting

• US 8,027,537 B1
• Filed: 05/18/2007
• Issued: 09/27/2011
• Est. Priority Date: 05/18/2007
• Status: Expired due to Fees

• Current Assignee: The United States of America As Represented By The Secretary of Agriculture
• Original Assignee: United States Of America As Represented By The Secretary Of The Air Force
• Inventors: Sorkin, Robert D., Repperger, Daniel W., Pinkus, Alan R.
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US11/805,262
• Time in Patent Office: 1,593 Days
• Field of Search: 382/181, 382/190, 382/224
• US Class Current: 382/181
• CPC Class Codes: G06F 18/253 of extracted features
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/806 of extracted features

Abstract

A method of distinguishing an object in an image. The method including the steps of detecting an image with an optical device where the image includes at least one object. Identifying at least one object using a central processing unit (CPU) that includes a central processing algorithm (CPA) and uses a majority voting analysis of multiple inputs to analyze at lest one object with the CPA. The image selected by the CPU is then displayed on the optical device.

Claims

1. A method for distinguishing an object in an image comprising the steps of

detecting an image with an optical device, the image including at least one object;

identifying at least one object using a central processing unit (CPU),

the CPU including a central processing algorithm (CPA),

the CPA includes a majority voting analysis of multiple inputs analyzing at least one object using the CPA;

displaying a CPU selected image on the optical device; and

,

wherein the CPA includes only three parsimonious information-theoretic measures.

2. The method of claim 1 wherein the three parsimonious information-theoretic measures are D_R, I(x;

y) and E_F.

3. A method for distinguishing an object in an image comprising the steps of

detecting an image with an optical device, the image including at least one object;

identifying at least one object using a central processing unit (CPU),

the CPU including a central processing algorithm (CPA),

the CPA includes a majority voting analysis of multiple inputs analyzing at least one object using the CPA;

displaying a CPU selected image on the optical device; and

,

wherein the multiple inputs are three independent inputs, the independent inputs are D_R, E_F, and I(x;

y).

4. A method for distinguishing an object in an image comprising the steps of

detecting an image with an optical device, the image including at least one object;

identifying at least one object using a central processing unit (CPU),

the CPU including a central processing algorithm (CPA),

the CPA includes a majority voting analysis of only three parsimonious information-theoretic measures to identify at least one object, the three parsimonious information-theoretic measures being D_R, I(x;

y) and E_F multiple inputs;

using the voting analysis of the three parsimonious information-theoretic measures to identify at least one object;

analyzing at least one object using the CPA,

displaying an identified object image on the optical device.

5. The method of claim 4 wherein the identified object image is a file image of the identified object.

6. The method of claim 4 wherein the identified object image is a text description of the identified object.

7. The method of claim 4 wherein the identified object image includes an identification certainty percentage.

8. The method of claim 4 wherein the optical device is any one of a binocular, a monocular, a night vision goggle, a helmet mounted display, or a combination thereof.

9. An optical apparatus for identifying an object in an image viewed by the optical apparatus, the optical apparatus including:

an optical viewer;

an objects selector for selecting an object to be identified;

a central processing unit for assessing the identity of the object to be identified, the assessing based upon only three independent inputs, the inputs including D_R, I(x;

y) and E_F.

10. The apparatus of claim 9 wherein the optical viewer is any one of a binocular, a monocular, a night vision goggle, a helmet mounted display, an aircraft heads up display, or a combination thereof.

11. The apparatus of claim 9 wherein the objects selector is a button to begin analysis on the object and a center of the optical viewer.

12. The apparatus of claim 9 wherein a representation of the object to be identified is displayed upon the optical viewer.

13. The apparatus of claim 9 wherein the identity of the object includes a degree of certainty of the identification of the object.