IDENTIFYING CHARACTER INFORMATION IN MEDIA CONTENT

• US 20080310722A1
• Filed: 06/15/2007
• Published: 12/18/2008
• Est. Priority Date: 06/15/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Kerth, Bradley, Daniels, Gary L.
• Granted Patent: US 7,929,764 B2
• Time in Patent Office: Days
• US Class Current: 382/182
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 30/158 using character size, text spacings or pitch estimation
• CPC Class Codes: G11B 27/10 Indexing; Addressing; Timing or synchronising; Measuring tape travel
• CPC Class Codes: G11B 27/3045 superimposed on the recorded main signal, e.g. burn-in-time code
• CPC Class Codes: H04N 21/43615 Interfacing a Home Network, e.g. for connecting the client to a plurality of peripherals home Audio Video Interoperability [HAVI] data switching networks H04L12/2805
• CPC Class Codes: H04N 21/44008 involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59
• CPC Class Codes: H04N 21/4425 Monitoring of client processing errors or hardware failure

Abstract

Implementations of identifying character information in media content are described. In one implementation, a frame of media content is marked with a frame identifier including one or more known characters. These known characters can uniquely identify the frame of media content. During transmission, compression, decompression, etc., of the frame, loss can occur. This loss can affect a quality of presentation of one or more of the known characters in the frame identifier. Therefore, when the frame is subsequently examined, the frame identifier can be identified, and best matches of known characters from a character recognition library can be found for characters in the frame identifier.

Claims

1. A method comprising:

recognizing at least one character of frame identification information in a frame of media content;

comparing the at least one character against a set of known characters in a character recognition library;

finding at least one known character from the set of known characters which best matches the at least one character, and

identifying the at least one known character as a frame identifier for the frame of media content.

2. A method as recited in claim 1, wherein the recognizing includes finding the at least one character at a known location in the frame of media content.

3. A method as recited in claim 1, further comprising calculating a quality of resolution of the at least one character based on a degree of similarity between the at least one character and the at least one known character.

4. A method as recited in claim 3, wherein the calculating includes computing the degree of similarity by dividing a total number of matching pixels between the at least one character and the at least one known character by a sum of the total number of matching pixels and a total number of nonmatching pixels between the at least one character and the at least one known character.

5. A method as recited in claim 3, further comprising estimating a quality of resolution of the frame of media content based on the quality of resolution of the at least one character.

6. A method as recited in claim 1, further comprising:

accessing a time of recordation of the frame of media content based on the frame identifier;

calculating an expected time of rendering of the frame of media content based on the time of recordation of the frame identifier; and

registering that the media content is being rendered incorrectly when a time at which the frame of media content is rendered differs from the expected time of rendering of the frame of media content.

7. A method as recited in claim 6, wherein the calculating includes:

identifying a second frame identifier associated with a second frame of media content, the second frame of media content being in a same file as the frame of media content;

accessing a time of recordation of the second frame of media content based on the second frame identifier;

computing an expected amount of time between a rendering of the second frame of media content and a rendering of the frame of media content based on the time of recordation of the second frame of media content and the time of recordation of the frame of media content; and

computing the expected time of rendering of the frame of media content as being a time of rendering of the second frame of media content plus the expected amount of time between the time of recordation of the second frame of media content and the time of recordation of the frame of media content.

8. A method as recited in claim 1, further comprising:

identifying a second frame identifier associated with a second frame of media content, wherein the second frame of media content is in a same file as the frame of media content, and further wherein a media function is activated at a time of rendering of the second frame of media content and deactivated at a time of rendering of the frame of media content;

registering that the media content is being rendered incorrectly if a time between the time of rendering of the second frame of media content and the time of rendering of the frame of media content is unequal to an accelerated elapsed time, wherein the accelerated elapsed time is calculated by dividing a difference between a time of recordation of the second frame based on the second frame identifier and a time of recordation of the frame based on the frame identifier, by an advance rate of the media function.

9. A method as recited in claim 1, further comprising:

determining a second frame identifier for a second frame of media content rendered immediately proximate to the frame of media content;

ascertaining that one or more frames of media content have been dropped if the second frame identifier is not immediately proximate in sequence to the frame identifier.

10. A computer-readable medium having computer-readable instructions residing thereon that, when executed, perform acts comprising:

locating frame identification information in a frame of media content;

estimating a frame identifier having one or more known characters by recognizing one or more characters in the frame identification information, and finding a best match between the one or more characters and a set of known characters in a character recognition library.

11. A computer-readable medium as recited in claim 10 having a set of computer-readable instructions that, when executed, perform acts further comprising calculating a quality of resolution of the frame identification information based on a degree of similarity between the one or more characters in the frame identification information and the one or more known characters in the frame identifier.

12. A computer-readable medium as recited in claim 11 having a set of computer-readable instructions that, when executed, perform acts further comprising estimating a quality of resolution of the frame of media content based on the quality of resolution of the frame identification information.

13. A computer-readable medium as recited in claim 10 having a set of computer-readable instructions that, when executed, perform acts further comprising evaluating a quality of resolution of the frame of media content by comparing the frame of media content with a control frame of media content having a same frame identifier as the frame identifier, wherein the frame of media content is based on the control frame of media content.

14. A computing-based device comprising:

a processor;

a memory including a quality sampling module comprising;

a character recognition module configured to;

recognize frame identification information including one or more characters in a frame of media content; and

identify a frame identifier including one or more known characters from among a set of known characters in a character representation library, wherein the one or more known characters are a best match between the one or more characters in the frame identification information and the set of known characters.

15. A computing-based device as recited in claim 14, wherein the character representation library resides in the memory.

16. A computing-based device as recited in claim 14, wherein the character representation library includes one or more of:

arabic numerals;

roman letters.

17. A computing-based device as recited in claim 14, wherein the quality sampling module further comprises a resolution quality module configured to compute a quality of resolution of the frame identification information by computing a similarity between the one or more known characters in the frame identifier and the one or more characters in the frame identification information.

18. A computing-based device as recited in claim 17, wherein the resolution quality module is further configured to estimate a quality of resolution of the frame of media content based on the quality of resolution of the frame identification information.

19. A computing-based device as recited in claim 14, wherein the quality sampling module further comprises a playback calibration module configured to determine if the frame of media content is being rendered correctly by comparing an expected presentation time associated with the frame identifier with an actual presentation time associated with the frame identifier.

20. A computing-based device as recited in claim 14, wherein the quality sampling module further comprises a playback calibration module configured to determine that audio content and video content in the media content are being rendered in synch if an actual presentation time of a frame of video content is the same as an actual rendering time of a portion of audio content, wherein the frame of video content has a frame identifier indicating that the frame of video content is to be rendered at the same time as the portion of audio content.