Two dimensional autonomous isotropic detection technique

• US 20030072470A1
• Filed: 10/15/2001
• Published: 04/17/2003
• Est. Priority Date: 10/15/2001
• Status: Active Grant

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Lee, Henry C.
• Granted Patent: US 7,298,866 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/10044 Radar image
• CPC Class Codes: G06T 2207/30212 Military
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/155 involving morphological operators
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

The present invention discloses a target detection process that acquires imagery from a target; compares the acquired imagery with image metric data; applies criteria to eliminate false detections and reduce clutter; applies morphological operators on the acquired imagery; probability rank orders the target imagery; divides the imagery into a series of detected target windows; and displays the detected target windows. The imagery undergoes light target detection, dark target detection or both. A morphological operator isolates targets rom their background. Two concatenated morphological filter patterns are used to screen imagery data. Spatial discontinuities at the pixel level can be detected. The detected target window images are presented to a user in a mosaic format.

Claims

1. A method for detecting an object image within image data comprising:

receiving image data;

segmenting the image data into multiple windows;

determining a likelihood that each window contains the object, and probability rank ordering the multiple windows based on the step of determining; and

selecting a predetermined one of the multiple windows as a window wherein the object image is considered to reside.

2. The method of claim 1 wherein the receiving step comprises:

collecting and recording the image data as the data emanates back to a receiver.

3. The process of claim 1, wherein the step of segmenting comprises:

determining a set of image metric data;

applying selection criteria to filter false detections and clutter from the image data;

comparing image data, after applying the selection criteria, with the image metric data; and

applying morphological operators on the image data.

4. The process of claim 1, comprising:

displaying at least one of the multiple windows.

5. The process of claim 2, comprising:

identifying pixels having a lighter contrast compared to other pixels in the imagery.

6. The process of claim 2, comprising:

identifying pixels having a darker contrast compared to other pixels in the imagery.

7. The process of claim 2, comprising:

identifying pixels having both lighter and darker contrast compared to other pixels in the imagery.

8. The process of claim 2, comprising:

using a morphological operator to isolate targets from their background.

9. The process of claim 2, comprising:

filtering the image data using two concatenated morphological filters.

10. The process of claim 2, comprising:

detecting spatial discontinuities at a pixel level.

11. The process of claim 2, comprising:

presenting the image data of multiple windows on a display in a mosaic format.

12. The process of claim 2, comprising:

communicating the detected window images to another system.

13. The process of claim 2, comprising:

the processing of image data comprising visual data.

14. The process of claim 2, comprising:

the processing of image data comprising thermal data.

15. The process of claim 2, comprising:

the processing of image data comprising synthetic aperture radar (SAR) data.

17. The process of claim 15, comprising:

a decision making authority that extracts windows and rank orders them.

18. The process of claim 15, comprising:

an image that is down-sampled n-times using a series of low pass filters that can filter in a horizontal and vertical direction.

19. The process of claim 15, comprising:

an image that has been down-sampled n-times, where n comprises a large number that can still accomplish target detection after accomplishing a larger amount of down-sampling.

20. The process of claim 15, comprising:

a filtering process performed by a six by six (6×

6) convolution filter.

21. The process of claim 15, comprising:

a filtering process performed by an N by N convolution filter, where N is a number greater than or equal to one.

16. A target detection process comprising:

acquiring image data;

down-sampling the image data n-times;

processing the down-sampled image data for detecting at least one of a light target and a dark target;

labeling subsets of the image data that may contain target data and rejecting clutter associated with these subsets of the image data;

combining results of the image data that has been down-sampled; and

forwarding combined results to a decision making authority.