System and method for improved optical character recognition for automated set-top box testing

• US 9,942,543 B2
• Filed: 10/31/2016
• Issued: 04/10/2018
• Est. Priority Date: 11/26/2010
• Status: Active Grant

• Current Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Original Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Inventors: Friel, Liam
• Primary Examiner(s): ALAM, MUSHFIKH I
• Application Number:
  US15/338,983
• Publication Number:
  US 20170048519A1
• Time in Patent Office: 526 Days
• Field of Search: 725107
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 20/41 Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items segmenting video sequences G06V20/49
• CPC Class Codes: H04N 17/00 Diagnosis, testing or measuring for television systems or their details
• CPC Class Codes: H04N 17/04 for receivers
• CPC Class Codes: H04N 17/045 Self-contained testing apparatus

Abstract

The present application provides a user configurable test system for set-top boxes (STB) and other consumer devices providing video output. In particular, it provides for a method of improving an Optical Character Recognition (OCR) process in such test systems.

Claims

1. A first device comprising:

one or more processors to;

obtain a video frame, comprising images in sequence, from a second device under test;

select a filter configuration, of a plurality of filter configurations, for at least one filter;

filter an image included in the images in sequence using the filter configuration to generate a filtered image;

identify text from the filtered image;

compare the text identified from the filtered image and an expected result;

determine a filter performance for the filtered image based on comparing the text identified from the filtered image and the expected result; and

determine a suitable filter configuration based on determining the filter performance.

2. The first device of claim 1, where the one or more processors, when filtering the image, are to one or more of:

remove one or more color components from the image,

adjust a contrast of the image,

invert colors in the image,

blur the image,

sharpen the image, or

zoom in on the image.

3. The first device of claim 1, where the one or more processors, when comparing the text, are to:

provide the filtered image to an optical character recognition engine to determine a success of comparing the text identified from the filtered image and the expected result.

4. The first device of claim 1, where the one or more processors are further to:

determine that a result is not the expected result based on comparing the text;

store the result for reference; and

utilize the result subsequently in a test routine.

5. The first device of claim 1, where each of the images in sequence includes a same text content.

6. The first device of claim 1, where the one or more processors, when comparing the text, are to:

compare a part of the text identified from the filtered image and the expected result,

the part of the text being less than an entirety of the text.

7. The first device of claim 1, where the one or more processors are further to:

test an accuracy of each filter configuration of the plurality of filter configurations; and

where the one or more processors, when determining the suitable filter configuration, are to;

determine the suitable filter configuration based on testing the accuracy of each filter configuration of the plurality of filter configurations.

8. A non-transitory computer-readable medium storing instructions, the instructions comprising:

one or more instructions that, when executed by one or more processors, cause the one or more processors to;

obtain a video frame, comprising a plurality of images, from a device under test;

select a filter configuration for a plurality of filters;

filter an image included in the plurality of images using the filter configuration to generate a filtered image;

identify text from the filtered image;

compare the text identified from the filtered image and an expected result;

determine a filter performance for the filtered image based on comparing the text identified from the filtered image and the expected result; and

determine a suitable filter configuration based on determining the filter performance.

9. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to determine the filter performance, further cause the one or more processors to:

determine that the filter performance is a positive result; and

filter another image included in the plurality of images using the filter configuration based on determining that the filter performance is the positive result.

10. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to determine the filter performance, further cause the one or more processors to:

determine that the filter performance is a negative result; and

repeatedly select a different filter configuration and filter the image using the different filter configuration until a positive result is determined.

11. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to obtain the video frame from the device under test, cause the one or more processors to:

obtain the video frame such that the plurality of images are chosen at intervals apart.

12. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to filter the image, further cause the one or more processors to:

remove one or more color components from the image,

adjust a contrast of the image,

invert colors in the image,

blur the image,

sharpen the image, or

zoom in on the image.

13. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to compare the text, further cause the one or more processors to:

provide the filtered image to an optical character recognition engine to determine a success of comparing the text identified from the filtered image and the expected result.

14. The non-transitory computer-readable medium of claim 8, where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:

determine that a result is not the expected result based on comparing the text;

store the result for reference; and

utilize the result subsequently in a test routine.

15. A method comprising:

obtaining, by a first device, a video frame, comprising images in sequence, from a second device under test;

selecting, by the first device, a filter configuration, of a plurality of filter configurations, for one or more filters;

filtering, by the first device, an image included in the images in sequence using the filter configuration to generate a filtered image;

identifying, by the first device, text from the filtered image;

comparing, by the first device, the text identified from the filtered image and an expected result;

determining, by the first device, a filter performance for the image based on comparing the text identified from the filtered image to the expected result; and

determining, by the first device, a suitable filter configuration based on determining the filter performance.

16. The method of claim 15, where each of the images in sequence includes a same text context.

17. The method of claim 15, where comparing the text comprises:

comparing a part of the text identified from the filtered image to the expected result,

the part of the text being less than an entirety of the text.

18. The method of claim 15, further comprising:

testing an accuracy of each filter configuration of the plurality of filter configurations; and

where determining the suitable filter configuration comprises;

determining the suitable filter configuration based on testing the accuracy of each filter configuration of the plurality of filter configurations.

19. The method of claim 15, where determining the filter performance comprises:

determining the filter performance indicates a positive result; and

filtering another image included in the images in sequence using the filter configuration based on determining that the filter performance indicates the positive result.

20. The method of claim 15, where determining the filter performance comprises:

determining the filter performance indicates a negative result; and

repeatedly selecting a different filter configuration and filtering the image using the different filter configuration until a positive result is indicated.