SYSTEM AND METHOD FOR GLOBAL OBJECT RECOGNITION

• US 20160378789A1
• Filed: 07/27/2015
• Published: 12/29/2016
• Est. Priority Date: 07/25/2014
• Status: Abandoned Application

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Hershey, Paul C., Sica, Michael P., Dudash, Jason, Graham, Christopher J.
• Application Number:
  US14/809,352
• Publication Number:
  US 20160378789A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/5854 using shape and object relationship

Abstract

A global object detection server reduces the amount of time needed to determine whether an object is present in a collection of images for a geographic area. In particular, the disclosed global object detection server selects one or more object recognition algorithms from a collection of algorithms based on one or more characteristics of the object to be detected. The algorithm results may then be fed back to reduce input data sets from iterative collections for similar regions. The global object detection server can also derive stochastic probabilities for object detection accuracy. Thereafter, one or more visualizations may be created that show confidence levels for the object'"'"'s probable location in the collection of images.

Claims

1. A computer-implemented method for global object recognition comprising:

receiving, by one or more hardware processors, a plurality of context parameters that define a geographic search for an object to be located;

receiving, by the one or more hardware processors, a plurality of characteristics that define the object to be located;

retrieving, based on the plurality of context parameters, one or more images representing a geographic location;

selecting, from a plurality of image processing algorithms, a subset of image processing algorithms to be used in processing the retrieved one or more images;

processing the retrieved one or more images according to the selected subset of image processing algorithms to obtain a plurality of results, at least one result indicating whether the object was identified in a corresponding image; and

determining at least one confidence value representing whether the object was located in one or more of the retrieved one or more images based on the at least one result.

2. The computer-implemented method of claim 1, wherein the subset of image processing algorithms are selected based on the context parameters that define the geographic search.

3. The computer-implemented method of claim 1, wherein the subset of image processing algorithms are selected based on the plurality of characteristics that define the object to be located.

4. The computer-implemented method of claim 1, wherein determining the at least one confidence value includes applying a stochastic processor to the at least one result.

5. The computer-implemented method of claim 1, wherein the one or more images comprise a first plurality of images representing the geographic location and a second plurality of images representing the geographic location;

and

the method further comprises removing the second plurality of images from the one or more images based on having processed the first plurality of images according to the selected subset of image processing algorithms.

6. The computer-implemented method of claim 1, further comprising:

generating a visualization of the at least one confidence value mapped to the geographic location corresponding to the at least one image.

7. The computer-implemented method of claim 1, further comprising:

implementing at least one image processing algorithm from the subset of image processing algorithms on at least one network node of a network node cluster in communication with the one or more hardware processors, the at least one network node configured to execute the at least one image processing algorithm on at least one image of the retrieved one or more images.

11. The system of claim 1, wherein determining the at least one confidence value includes applying a stochastic processor to the at least one result.

12. The system of claim 1, wherein the one or more images comprise a first plurality of images representing the geographic location and a second plurality of images representing the geographic location;

and

the plurality of operations further comprise removing the second plurality of images from the one or more images based on having processed the first plurality of images according to the selected subset of image processing algorithms.

13. The system of claim 1, wherein the plurality of operations further comprise generating a visualization of the at least one confidence value mapped to the geographic location corresponding to the at least one image.

14. The system of claim 1, wherein the plurality of operations further comprise:

implementing at least one image processing algorithm from the subset of image processing algorithms on at least one network node of a network node cluster in communication with the at least one hardware processor, the at least one network node configured to execute the at least one image processing algorithm on at least one image of the retrieved one or more images.

8. A system for global object recognition comprising:

a machine-readable medium storing computer-executable instructions; and

at least one hardware processor in communication with the machine-readable medium that, having executed the computer-executable instructions, performs a plurality of operations, the operations comprising;

receiving, by one or more hardware processors, a plurality of context parameters that define a geographic search for an object to be located;

receiving, by the one or more hardware processors, a plurality of characteristics that define the object to be located;

retrieving, based on the plurality of context parameters, one or more images representing a geographic location;

selecting, from a plurality of image processing algorithms, a subset of image processing algorithms to be used in processing the retrieved one or more images;

processing the retrieved one or more images according to the selected subset of image processing algorithms to obtain a plurality of results, at least one result indicating whether the object was identified in a corresponding image; and

determining at least one confidence value representing whether the object was located in one or more of the retrieved one or more images based on the at least one result.

9. The system of claim 8, wherein the at least one hardware processor selects the subset of image processing algorithms based on the context parameters that define the geographic search.

10. The system of claim 8, wherein the at least one hardware processor selects the subset of image processing algorithms on the plurality of characteristics that define the object to be located.

15. A machine-readable medium having computer-executable instructions stored thereon that, when executed by at least one hardware processor, cause the at least one hardware processor to:

receive a plurality of context parameters that define a geographic search for an object to be located;

receive a plurality of characteristics that define the object to be located;

receive, based on the plurality of context parameters, one or more images representing a geographic location;

select, from a plurality of image processing algorithms, a subset of image processing algorithms to be used in processing the retrieved one or more images;

process the retrieved one or more images according to the selected subset of image processing algorithms to obtain a plurality of results, at least one result indicating whether the object was identified in a corresponding image; and

determine at least one confidence value representing whether the object was located in one or more of the retrieved one or more images based on the at least one result.

16. The machine-readable medium of claim 15, wherein the selection of the subset of image processing algorithms is based on the context parameters that define the geographic search.

17. The machine-readable medium of claim 15, wherein the selection of the subset of image processing algorithms is based on the plurality of characteristics that define the object to be located.

18. The machine-readable medium of claim 15, wherein the determination of the at least one confidence value includes applying a stochastic processor to the at least one result.

19. The machine-readable medium of claim 15, wherein the one or more images comprise a first plurality of images representing the geographic location and a second plurality of images representing the geographic location;

and

the at least one hardware processor further removes the second plurality of images from the one or more images based on having processed the first plurality of images according to the selected subset of image processing algorithms.

20. The machine-readable medium of claim 15, wherein the at least one hardware processor further generates a visualization of the at least one confidence value mapped to the geographic location corresponding to the at least one image.

21. A computer-implemented method for global object recognition comprising:

receiving, by one or more hardware processors, a plurality of context parameters that define a search for an object to be located;

receiving, by the one or more hardware processors, a plurality of characteristics that define the object to be located;

retrieving, based on the plurality of context parameters, a plurality of source data;

selecting, from a plurality of processing algorithms, a subset of processing algorithms to be used in processing the retrieved one or more source data;

processing the retrieved plurality of source data according to the selected subset of processing algorithms to obtain a plurality of results, at least one result indicating whether the object was identified in a corresponding source data; and

determining at least one confidence value representing whether the object was located in one or more of the retrieved source data based on the at least one result.

22. The computer-implemented method of claim 21, wherein the method further comprises building one or more algorithm chains indicating the order in which the algorithms are to be executed.

23. The computer-implemented method of claim 22, wherein the one or more algorithm chains are built to increase overall processing speed by reducing the source data.

24. The computer-implemented method of claim 23, wherein the method further comprises at each said algorithm in the chain removing one or more source data based on having processed the plurality of source data according to said algorithm.

25. The computer-implemented method of claim 24, wherein the source data is an image.