Heterogeneous image processing system

• US 8,326,092 B2
• Filed: 04/23/2007
• Issued: 12/04/2012
• Est. Priority Date: 04/23/2007
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Chung, William H., Kim, Moon J., Moulic, James R., Sanuki, Toshiyuki
• Primary Examiner(s): TSAI, TSUNG YIN
• Application Number:
  US11/738,723
• Publication Number:
  US 20080260297A1
• Time in Patent Office: 2,052 Days
• Field of Search: None
• US Class Current: 382/307
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining

Abstract

The present invention relates to machine vision computing environments, and more specifically relates to a system and method for selectively accelerating the execution of image processing applications using a hybrid computing system. To this extent, a hybrid system is generally defined as one that is multi-platform, and potentially distributed via a network or other connection. The invention provides a machine vision system and method for executing image processing applications on a hybrid image processing system referred to herein as an image co-processor that comprises (among other things) a plurality of special purpose engines (SPEs) that work to process multiple images in an accelerated fashion.

Claims

1. A heterogeneous image processing system, comprising:

an image co-processor comprising a plurality of distinct special purpose engines (SPEs), the image co-processor being configured to;

receive a plurality of images collectively and substantially simultaneously recorded by a plurality of image recordation mechanisms, each image recordation mechanism providing at least one of the plurality of images for acquisition by one of a plurality of image grabbers, each image grabber sending at least one of the plurality of images to one of a plurality of peripheral component interconnects, each of the plurality of peripheral component interconnects sending at least one of the plurality of images to a pure load balancer;

distribute distinct portions of the plurality of images across the plurality of SPEs, wherein the distinct portions are different views of an object recorded at substantially a same time by at least two of the plurality of image recordation mechanisms;

collectively and substantially simultaneously process the distinct portions of the plurality of images at the plurality of SPEs to determine associated image data; and

receive a selective off-load of at least one library function at a run time, wherein the at least one library function is written for a plurality of parallel multi-core processors, wherein the at least one library function supports the processing of the distinct portions of the plurality of images at the plurality of SPEs, and wherein the at least one library function is one of at least one algorithm, an optimization for a processor architecture, and a wrapper for a scripting language.

2. The heterogeneous image processing system of claim 1, further comprising a control processor being configured to receive the image data from the image co-processor, and to determine control information for the plurality of images.

3. The heterogeneous image processing system of claim 2, further comprising:

a staging storage unit for storing the image data for use by the control processor; and

an archive storage unit for storing at least one of the image data or the control information.

4. The heterogeneous image processing system of claim 1, further comprising an input/output (I/O) processor, the I/O processor comprising:

a set of express peripheral component interconnects (PCIs);

a pure load balancer (PLB) coupled to the set of express PCIs; and

a network interface coupled to the PLB for interfacing with at least one legacy application.

5. The heterogeneous image processing system of claim 4, the I/O processor receiving the plurality of images from an image grabber, and providing the plurality of images to the image co-processor.

6. The heterogeneous image processing system of claim 1, the image co-processor further comprising:

a power processing element (PPE); and

an element interconnect bus (EIB) coupled to the PPE and the plurality of SPEs.

7. The heterogeneous image processing system of claim 1, the SPEs comprising accelerators that increase a rate of processing of the plurality of images.

8. The heterogeneous image processing system of claim 1, the heterogeneous image processing system being operable with multiple different computing platforms.

9. The heterogeneous image processing system of claim 1, wherein the different views comprise different angles of view.

10. The heterogeneous image processing system of claim 1, wherein the different views comprise different perspectives of view.

11. A heterogeneous image processing method, comprising:

receiving a plurality of images in an image co-processor, the image co-processor comprising a plurality of special purpose engines (SPEs), wherein the plurality of images include images collectively and substantially simultaneously recorded by a plurality of image recordation mechanisms, each image recordation mechanism providing at least one of the plurality of images for acquisition by one of a plurality of image grabbers, each image grabber sending at least one of the plurality of images to one of a plurality of peripheral component interconnects, each of the plurality of peripheral component interconnects sending at least one of the plurality of images to a pure load balancer;

distributing distinct portions of the plurality of images across the plurality of SPEs, wherein the distinct portions are different views of an object recorded at substantially a same time by at least two of the plurality of image recordation mechanisms;

processing the distinct portions of the plurality of images collectively and substantially simultaneously with the plurality of SPEs to determine image data associated with the distinct portions of the plurality of images;

providing at least one of the image data or the distinct portions of the plurality of images to a control processor to determine control information associated with the distinct portions of the plurality of images; and

receiving a selective off-load of at least one library function at a run time, wherein the at least one library function is written for a plurality of parallel multi-core processors, wherein the at least one library function supports the processing of the distinct portions of the plurality of images at the plurality of SPEs, and wherein the at least one library function is one of at least one algorithm, an optimization for a processor architecture, and a wrapper for a scripting language.

12. The heterogeneous image processing method of claim 11, further comprising:

determining, the control information with the control processor.

13. The heterogeneous image processing method of claim 12, further comprising:

storing the image data in a staging storage unit for processing by the control processor to determine control information; and

storing at least one of the image data or the control information in an archive storage unit.

14. The heterogeneous image processing method of claim 11, further comprising:

receiving the plurality of images in an input/output (I/O) processor, the I/O processor comprising;

a set of express peripheral component interconnects (PCIs);

a pure load balancer (PLB) coupled to the set of express PCIs; and

a network interface coupled to the PLB for interfacing with at least one legacy application.

15. The heterogeneous image processing method of claim 14, the I/O processor receiving the plurality of images from an image grabber, and providing the plurality of images to the plurality of SPEs.

16. The heterogeneous image processing method of claim 11, the image co-processor further comprising:

a power processing element (PPE); and

an element interconnect bus (EIB) coupled to the PPE and the plurality of SPEs.

17. The heterogeneous image processing method of claim 11, the plurality of SPEs comprising accelerators that increase a rate of processing of the plurality of images.

18. The heterogeneous image processing method of claim 11, further comprising receiving the plurality of images in an image grabber from at least one image recording devices.

19. A program product stored on computer readable storage medium for heterogeneously processing an image, the computer readable storage medium comprising program code for causing a computer system to:

receive a plurality of images in an image co-processor, the image co-processor comprising a plurality of special purpose engines (SPEs), wherein the plurality of images include images collectively and substantially simultaneously recorded by a plurality of image recordation mechanisms, each image recordation mechanism providing at least one of the plurality of images for acquisition by one of a plurality of image grabbers, each image grabber sending at least one of the plurality of images to one of a plurality of peripheral component interconnects, each of the plurality of peripheral component interconnects sending at least one of the plurality of images to a pure load balancer;

distribute distinct portions of the plurality of images across the plurality of SPEs, wherein the distinct portions are different views of an object recorded at substantially a same time by at least two of the plurality of image recordation mechanisms;

process the distinct portions of the plurality of images collectively and substantially simultaneously with the plurality of SPEs to determine image data associated with the plurality of images;

provide at least one of the image data or the distinct portions of the plurality of images to a control processor to determine control information associated with the distinct portions of the plurality of images; and

receive a selective off-load of at least one library function at a run time, wherein the at least one library function is written for a plurality of parallel multi-core processors, wherein the at least one library function supports the processing of the distinct portions of the plurality of images at the plurality of SPEs, and wherein the at least one library function is one of at least one algorithm, an optimization for a processor architecture, and a wrapper for a scripting language.

20. The program product of claim 19, the computer readable storage medium further comprising program code for causing the computer system to:

determine control information for the plurality of images with the control processor.

21. The program product of claim 20, the computer readable storage medium further comprising program code for causing the computer system to:

store the image data in a staging storage unit for processing by the control processor; and

store the at least one of the image data or the plurality of images in an archive storage unit.

22. The program product of claim 19, the computer readable storage medium further comprising program code for causing the computer system to:

receive the plurality of images in an input/output (I/O) processor, the I/O processor comprising;

a set of express peripheral component interconnects (PCIs);

a pure load balancer (PLB) coupled to the set of express PCIs; and

a network interface coupled to the PLB for interfacing with at least one legacy application.

23. The program product of claim 22, the I/O processor receiving the plurality of images from an image grabber, and providing the plurality of images to the plurality of SPEs.

24. The program product of claim 19, the image co-processor further comprising:

a power processing element (PPE); and

an element interconnect bus (EIB) coupled to the PPE and the plurality of SPEs.

25. The program product of claim 19, the SPEs comprising accelerators that increase a rate of processing of the plurality of images.

26. The program product of claim 19, further comprising receiving the plurality of images in an image grabber from at least one image recording devices.

27. A method for deploying a system for heterogeneously processing an image, comprising:

providing a computer infrastructure being operable to;

receive a plurality of images in an image co-processor, the image co-processor comprising a plurality of special purpose engines (SPEs), wherein the plurality of images include images collectively and substantially simultaneously recorded by a plurality of image recordation mechanisms, each image recordation mechanism providing at least one of the plurality of images for acquisition by one of a plurality of image grabbers, each image grabber sending at least one of the plurality of images to one of a plurality of peripheral component interconnects, each of the plurality of peripheral component interconnects sending at least one of the plurality of images to a pure load balancer;

distribute distinct portions of the plurality of images across the plurality of SPEs, wherein the distinct portions are different views of an object recorded at substantially a same time by at least two of the plurality of image recordation mechanisms;

process the distinct portions of the plurality of images collectively and substantially simultaneously with the plurality of SPEs to determine image data associated with the distinct portions of the plurality of images; and

receive a selective off-load of at least one library function at a run time, wherein the at least one library function is written for a plurality of parallel multi-core processors, wherein the at least one library function supports the processing of the distinct portions of the plurality of images at the plurality of SPEs, and wherein the at least one library function is one of at least one algorithm, an optimization for a processor architecture, and a wrapper for a scripting language.