System and method for capturing and detecting symbology features and parameters

• US 9,189,670 B2
• Filed: 02/09/2010
• Issued: 11/17/2015
• Est. Priority Date: 02/11/2009
• Status: Active Grant

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Moed, Michael C., Akopyan, Mikhail
• Primary Examiner(s): JOHNSON, SONJI N
• Application Number:
  US12/703,077
• Publication Number:
  US 20100200660A1
• Time in Patent Office: 2,107 Days
• US Class Current: 1/1
• CPC Class Codes: G06K 7/14 using light without selection of wavelength, e.g. sensing reflected white light G06K7/10831 - G06K7/1097 take precedence

Abstract

This invention provides a system and method for capturing, detecting and extracting features of an ID, such as a 1D barcode, that employs an efficient processing system based upon a CPU-controlled vision system on a chip (VSoC) architecture, which illustratively provides a linear array processor (LAP) constructed with a single instruction multiple data (SIMD) architecture in which each pixel of the rows of the pixel array are directed to individual processors in a similarly wide array. The pixel data are processed in a front end (FE) process that performs rough finding and tracking of regions of interest (ROIs) that potentially contain ID-like features. The ROI-finding process occurs in two parts so as to optimize the efficiency of the LAP in neighborhood operations—a row-processing step that occurs during image pixel readout from the pixel array and an image-processing step that occurs typically after readout occurs. The relative motion of the ID-containing ROI with respect to the pixel array is tracked and predicted. An optional back end (BE) process employs the predicted ROI to perform feature-extraction after image capture. The feature extraction derives candidate ID features that are verified by a verification step that confirms the ID, creates a refined ROI, angle of orientation and feature set. These are transmitted to a decoding processor or other device.

Claims

1. A system for detecting and extracting symbology ID features comprising:

a. a plurality of single-instruction, multiple-data (SIMD) processors, a memory, and at least one general purpose processor all residing on a discrete die, the memory being constructed and arranged to store and transfer, and the SIMD processors each being constructed and arranged to operate on, a group of pixel data provided by a pixel array; and

b. an ID region of interest (ROI) finding process that detects at least one ID ROI that contain ID characteristics and that operates at least in part while captured pixel data associated with the ID ROI is provided from the pixel array so as to provide a data set that is transmitted to an ID decoding process;

wherein the ROI-finding process includes a row-processing process that operates within a row to detect edge data indicating an ID ROI candidate while captured pixel data is read-out from the pixel array.

2. The system as set forth in claim 1 wherein the pixel array is on the die.

3. The system as set forth in claim 1 wherein each of the SIMD processors is constructed and arranged to operate on the group of pixel data provided by the pixel array on a common clock cycle.

4. The system as set forth in claim 1 wherein the group of pixel data comprises a row of pixel data read-out from the pixel array.

5. The system as set forth in claim 1 wherein the ROI-finding process includes an image-processing process that operates to group each ID ROI candidate edge data into similar bins relative to orientations thereof.

6. The system as set forth in claim 5 wherein the image-processing process defines bins with maximum count of edge data therein.

7. The system as set forth in claim 5 wherein the image-processing process includes a blob analysis process that analyzes for edge data meeting ID size criteria and a process that generates at least one ROI based upon a mapping of analyzed ROI data that meets the ID size criteria.

8. The system as set forth in claim 1 further comprising an ROI tracking process that employs at least one ROI to define ROI motion characteristics so as to provide a predicted ROI.

9. The system as set forth in claim 8 further comprising a data refinement process that receives the predicted ROI and, based upon the predicted ROI and subsequent image capture, generates ID features within the ROI for decoding of information therein.

10. The system as set forth in claim 9 wherein the data refinement process further includes verification of ID information within the ID features, including further refinement of an angular orientation of the ID features.

11. A system for detecting and extracting symbology ID features comprising:

a. a plurality of single-instruction, multiple-data (SIMD) processors, a memory, and at least one general purpose processor all residing on a discrete die, the memory being constructed and arranged to store and transfer, and the SIMD processors each being constructed and arranged to operate on, a group of pixel data provided by a pixel array;

b. a front end (FE) process that determines a presence of ID characteristics within a grouping of pixel data while the pixel data is being captured and read from the pixel array, wherein the FE process includes an image-processing process that operates to group each ID ROI candidate edge data into similar bins relative to orientations of the ID ROI candidate edge data with respect to the pixel array; and

c. a transmission process that forwards information associated with the presence of the ID characteristics so as to allow a decoding process to decode at least a portion of the pixel data to derive ID information therefrom.

12. The system as set forth in claim 11 wherein pixel array is on the die.

13. The system as set forth in claim 11 wherein the decoding process is located off-die and operatively connected to the die.

14. The system as set forth in claim 11 further comprising an ROI motion-determination process that derives a predicted ROI.

15. The system as set forth in claim 14 wherein the predicted ROI information is transmitted to the decoding process in association with a stream of the pixel data read from the pixel array.

16. The system as set forth in claim 11 further comprising a back end (BE) process that captures the pixel data and extracts ID features therefrom and verifies presence of the ID features based upon the predicted region of interest, the extracted ID features being transmitted to the decoding process.

17. The system as set forth in claim 11 wherein the decoding process receives the presence of the ID characteristics as a trigger and in response thereto is constructed and arranged to decode an ID from a transmitted stream of the pixel data.

18. A method for detecting and extracting symbology ID features comprising the steps of:

a. with a plurality of single-instruction, multiple-data (SIMD) processors, a memory, and at least one general purpose processor all residing on a discrete die, storing and transferring, by the memory, a group of pixel data provided by a pixel array and operating, by at least one of the SIMD processors, on the group of pixel data; and

b. operating an ID region of interest (ROI) finding process that detects ID ROIs that contain ID characteristics and at least in part while captured pixel data associated with the ID ROI is provided from the pixel array so as to provide a data set that is transmitted to an ID decoding process;

c. operating a row-processing process within a row to detect edge data indicating an ID ROI candidate while captured pixel data is read-out from the pixel array.

19. The method as set forth in claim 18 further comprising locating the pixel array on the die.

20. The method as set forth in claim 18 wherein the step of operating by the at least one of the SIMD processors operates on the group of pixel data in a common clock cycle.

21. The method as set forth in claim 18 wherein the group of pixel data comprises a row of pixel data read-out from the pixel array.

22. The method as set forth in claim 18 further comprising operating an image-processing process that operates to group each ID ROI candidate edge data into similar bins relative to orientations thereof.

23. The method as set forth in claim 22 further comprising defining, in the image-processing process, bins with maximum count of edge data therein.

24. The method as set forth in claim 22 further comprising performing, in the image processing process, a blob analysis process that analyzes for edge data meeting ID size criteria and a process that generates at least one ROI based upon a mapping of analyzed ROI data that meets the ID size criteria.

25. The method as set forth in claim 18 further comprising tracking an ROI by employing at least one ROI to define ROI motion characteristics so as to provide a predicted ROI.

26. The method as set forth in claim 25 further comprising operating a data refinement process that receives the predicted ROI and, based upon the predicted ROI and subsequent image capture, generating ID features within the ROI for decoding of ID information therein.

27. The system as set forth in claim 26 further comprising verifying ID information within the ID features, including further refining of an angular orientation of the ID features.

28. A method for detecting and extracting symbology ID features comprising the steps of:

a. with a plurality of single-instruction, multiple-data (SIMD) processors, a memory, and at least one general purpose processor all residing on a discrete die, storing and transferring, by the memory, a group of pixel data provided by a pixel array and operating, by at least one of the SIMD processors, on the group of pixel data;

b. determining, with a front end (FE) process, a presence of ID characteristics within a grouping of pixel data while the pixel data is being captured and read from the pixel array, wherein the FE process includes an image-processing process that operates to group each ID ROI candidate edge data into similar bins relative to orientations of the ID ROI candidate edge data with respect to the pixel array; and

c. forwarding, with a transmission process, information associated with the presence of the ID characteristics so as to allow a decoding of at least a portion of the pixel data to derive ID information therefrom.

29. The method as set forth in claim 28 further comprising locating the pixel array on the die.

30. The method as set forth in claim 28 wherein the step of decoding is performed off-die.

31. The method as set forth in claim 28 further comprising deriving a predicted ROI with an ROI motion-determination process.

32. The method as set forth in claim 31 further comprising transmitting information of the predicted ROI to the decoding process in association with a stream of the pixel data read from the pixel array.

33. The method as set forth in claim 31 further comprising capturing the pixel data with a back end (BE) and extracting ID features therefrom.

34. The method as set forth in claim 33 further comprising verifying presence of the ID features based upon the predicted region of interest, the extracted ID features being transmitted to the decoding process.

35. The system as set forth in claim 28 the step of decoding includes receiving the presence of the ID characteristics as a trigger and in response thereto decoding an ID from a transmitted stream of the pixel data.