Vision sensors, systems, and methods

• US 9,451,142 B2
• Filed: 07/31/2008
• Issued: 09/20/2016
• Est. Priority Date: 11/30/2007
• Status: Active Grant

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Plummer, A. Piers N., McGarry, E. John
• Primary Examiner(s): HO, TUAN V
• Application Number:
  US12/184,187
• Publication Number:
  US 20090141987A1
• Time in Patent Office: 2,973 Days
• Field of Search: 348/222.1, 348/308, 348/86, 348/294, 382/190, 382/195, 382/153, 382/312, 382/218, 382/219, 250/208.1
• US Class Current: 1/1
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: G06V 10/955 using specific electronic processors
• CPC Class Codes: H04N 25/75 Circuitry for providing, modifying or processing image signals from the pixel array

Abstract

A single chip vision sensor of an embodiment includes a pixel array and one or more circuits. The one or more circuits are configured to search an image for one or more features using a model of the one or more features. A method of an embodiment in a single chip vision sensor includes obtaining an image based at least partially on sensed light, and searching the image for one or more features using a model of the one or more features. A system of an embodiment includes the single chip vision sensor and a device. The device is configured to receive one or more signals from the single chip vision sensor and to control an operation based at least partially on the one or more signals.

Claims

1. A single chip vision sensor, comprising:

a pixel array; and

one or more circuits configured to search an image for one or more features using a model of the one or more features, to record one or more poses of the model relative to the image that result in at least a partial match between the model and the image, and to transmit data of the image only if at least one of the one or more features are found in the image;

wherein said one or more circuits comprising a plurality of comparators for comparing, concurrently with each other, one or more corresponding values obtained based on data of the image with one or more values of the model.

2. The single chip vision sensor of claim 1,

said one or more circuits configured to process one or more signals provided from the pixel array to generate said image.

3. The single chip vision sensor of claim 1,

said model including a plurality of model instructions that each include a corresponding coordinate with respect to a model reference point.

4. The single chip vision sensor of claim 1,

said one or more circuits configured to transform one or more coordinates of the model in accordance with one or more affine transform values.

5. The single chip vision sensor of claim 1,

said one or more circuits configured to compare the model, after transforming one or more coordinates of the model, with the image for each of a plurality of positionings of the model relative to the image concurrently.

6. The single chip vision sensor of claim 5,

said plurality of positionings including positionings of a reference point of the model at a plurality of locations in a row of the image.

7. The single chip vision sensor of claim 1,

said one or more circuits comprising an image address processor configured to transform coordinates of the model into transformed coordinates based at least partially on one or more transform values.

8. The single chip vision sensor of claim 7,

said one or more circuits further comprising a plurality of image memory devices for storing image data of the image and for outputting data concurrently with each other in a case where the plurality of image memory devices are accessed using at least one of said transformed coordinates.

9. The single chip vision sensor of claim 7,

said one or more circuits further comprising an image data selector configured to receive data of the image and to shift the data by an amount that is based on at least one of said transformed coordinates.

10. The single chip vision sensor of claim 9,

said one or more circuits further comprising a plurality of accumulators, each accumulator of the plurality of accumulators configured to receive an accumulation increment of the model and to selectively perform accumulation to accumulate the accumulation increment based on an output of a corresponding comparator of the plurality of comparators.

11. The single chip vision sensor of claim 10,

said one or more circuits further comprising a model data shifter for shifting the one or more values and the accumulation increment of the model within the model data shifter and for providing the one or more values to the plurality of comparators and the accumulation increment to the plurality of accumulators.

12. A method in a single chip vision sensor, comprising:

obtaining an image based at least partially on sensed light;

searching the image for one or more features using a model of the one or more features;

recording one or more poses of the model relative to the image that result in at least a partial match between the model and the image;

transmitting data of the image only if at least one of the one or more features are found in the image; and

outputting information based on a result of said searching.

13. The method of claim 12, said searching comprising:

comparing the model, after transforming one or more coordinates of the model, with the image for each of a plurality of positionings of the model relative to the image concurrently.

14. The method of claim 13,

said plurality of positionings including positionings of a reference point of the model at a plurality of locations in a row of the image.

15. The method of claim 12, said searching comprising:

transforming coordinates of the model into transformed coordinates based on one or more transform values.

16. The method of claim 15, said searching further comprising:

causing a plurality of image memory devices to output data concurrently with each other by accessing the plurality of image memory devices using at least one of said transformed coordinates.

17. The method of claim 15, said searching further comprising:

shifting data of the image by an amount that is based on at least one of said transformed coordinates.

18. The method of claim 12, said searching comprising:

causing a plurality of comparators to compare, concurrently with each other, one or more corresponding values obtained based on data of the image with one or more values of the model.

19. The method of claim 18, said searching further comprising:

selectively performing accumulation to accumulate an accumulation increment in an accumulator based on an output of a corresponding comparator of the plurality of comparators.

20. A system, comprising:

a single chip vision sensor configured to obtain an image based at least partially on sensed light, to search the image for one or more features using a model of the one or more features, to record one or more poses of the model relative to the image that result in at least a partial match between the model and the image, and to transmit image data for regions of interest in the image that are determined based on identified features from the search, wherein said searching further comprising a plurality of comparators for comparing, concurrently with each other, one or more corresponding values obtained based on data of the image with one or more values of the model; and

a device for receiving one or more signals from the single chip vision sensor and for controlling an operation based at least partially on the one or more signals.

21. A single chip vision sensor, comprising:

a pixel array; and

one or more circuits configured to search an image for one or more features using a model of the one or more features, to record one or more poses of the model relative to the image that result in at least a partial match between the model and the image, and to transmit data of the image only if at least one of the one or more features are found in the image;

wherein said one or more circuits comprising a priority encoder for indicating positionings of the model in accordance with a binary tag word that is formed based on results of comparisons of the model, after one or more coordinates of the model have been transformed, with the image.