Target acquisition and tracking

• US 5,912,980 A
• Filed: 07/13/1995
• Issued: 06/15/1999
• Est. Priority Date: 07/13/1995
• Status: Expired due to Fees

• Current Assignee: Hunke H. Martin
• Original Assignee: Hunke H. Martin
• Inventors: Hunke, H. Martin
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US08/501,944
• Time in Patent Office: 1,433 Days
• Field of Search: 382/103, 382/164, 382/165, 382/169, 382/236, 348/169, 348/170, 348/171, 348/172, 348/416
• US Class Current: 382/103
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A method for automatically locating a predetermined target class of objects in a video image stream, comprising the steps of determining typical colors found in objects of the predetermined target class, detecting a moving area in the video image stream, determining the colors in the moving area, and determining whether the moving area contains colors similar to the predetermined target class typical colors. Additionally the method allows tracking of such located target objects in subsequent frames of the video image stream based upon the colors in the target objects.

Claims

1. A method for automatically locating a predetermined target class of objects in a video image stream, comprising the steps of:

(a) determining typical colors found in objects of said predetermined target class;

(b) detecting a moving area in said video image stream;

(c) determining the colors in said moving area;

(d) determining whether the moving area contains colors similar to the predetermined target class typical colors;

wherein said moving area contains at least a portion of a target object in said predetermined target class when said moving area contains colors similar to said predetermined target class typical colors.

2. The method recited in claim 1, further comprising the step of tracking said target object in subsequent frames of said video image stream based upon the colors in said target object.

3. The method recited in claim 1, further comprising the steps of:

(e) selecting a search area in a subsequent frame of said video image stream;

(f) determining colors in said search area;

(g) determining whether the search area colors are consistent with the colors in said target object.

4. The method recited in claim 3, further comprising the steps of predicting the location of said target object based upon the prior location of the said target object, wherein the predicted location is within said search area selected in step (e).

5. The method recited in claim 1, wherein a general target color classifier representative of said predetermined target class colors is used by step (d) to determine whether said moving area contains colors similar to said predetermined target class typical colors.

6. The method recited in claim 3, wherein a individual target color classifier representative of the target object colors is used by step (g) to determine whether said target object is in said search area.

7. The method recited in claim 5, further comprising the steps of modifying said general target color classifier in response to at least one operational condition.

8. The method recited in claim 7, wherein said at least one operational condition is lighting condition.

9. The method recited in claim 7, wherein said at least one operational condition are predetermined characteristics of equipment used to generate said video image stream.

10. The method recited in claim 6, further comprising the steps of modifying said individual target color classifier in response to at least one operational condition.

11. The method recited in claim 10, wherein said at least one operational condition is lighting condition.

12. The method recited in claim 5, wherein said general target color classifier is generated by sampling the colors of sample objects in said predetermined target class.

13. The method recited in claim 1, further comprising the steps of:

(e) ascertaining geometrical structures in said moving area; and

(f) comparing the moving area geometrical structures to the geometrical structures typical of said predetermined target class to determine if said moving area may contain said target object.