MOVING TEXT DETECTION IN VIDEO

• US 20100254605A1
• Filed: 04/01/2009
• Published: 10/07/2010
• Est. Priority Date: 04/01/2009
• Status: Active Grant

• Current Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Original Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Inventors: Wredenhagen, Gordon F.
• Granted Patent: US 8,718,366 B2
• Time in Patent Office: Days
• US Class Current: 382/176
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20044 Skeletonization; Medial axis transform
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• CPC Class Codes: H04N 5/144 Movement detection for video coding H04N19/503; analysis of motion in general G06T7/20

Abstract

Methods and apparatus for detecting moving text in video comprising receiving consecutive frames from a video stream, extracting a sequence of pixels from the consecutive frames, categorizing the pixels, thinning the pixels, correlating corresponding thinned pixels in the frames, identifying the peaks that are equal to or exceed a threshold, and performing further processing on the peaks to determine if the peaks contain moving text.

Claims

1. A method for detecting moving text in video comprising:

extracting a sequence of pixels from consecutive frames from a video stream; and

identifying peaks derived from pixels in the frames that are equal to or exceed a threshold.

2. The method of claim 1 further comprising:

categorizing the pixels;

thinning the pixels; and

correlating corresponding thinned pixels in the frames.

3. The method of claim 2 wherein the correlating is a phase plane correlation.

4. A method for detecting moving text in video comprising:

extracting consecutive frames from a video stream;

extracting rows of pixels from the consecutive frames;

extracting pixel values from the pixels; and

identifying peak candidates derived from the pixels that equal or exceed a threshold.

5. The method of claim 4 further comprising:

categorizing the pixels based on the extracted pixel values;

thinning the categorized pixels;

converting thinned categorized pixels; and

cross correlating corresponding thinned pixels from consecutive frames.

6. The method of claim 5 wherein the cross correlating is a phase plane correlation.

7. The method of claim 5 wherein the categorizing comprises:

subtracting a first pixel value from a second pixel value and obtaining a result; and

comparing the result to at least one threshold.

8. The method of claim 5 wherein thinning comprises determining a first instance of similarly categorized pixels.

9. A moving text detection device comprising:

an extractor configured to extract from a video stream any of a video frame, a row of pixels, a pixel, or a pixel value;

a processor configured to categorize pixels;

a thinning module configured to thin the categorized pixels; and

a correlator configured to perform correlation of the thinned pixels.

10. The moving text detection device of claim 9 further comprising:

a receiver configured to receive a video stream.

11. The moving text detection device of 9 further comprising:

a state machine configured to perform thinning or extraction.

12. The moving text detection device of 9 wherein the correlator is a phase plane correlator.

13. A moving text detection device comprising:

an extractor configured to extract consecutive frames from a video stream;

the extractor further configured to extract rows of pixels from the consecutive frames;

the extractor further configured to extract pixel values from the pixels; and

a processor configured to identify peak candidates from pixels that equal or exceed a threshold.

14. The moving text detection device of claim 13 further comprising:

the processor further configured to categorize the pixels based on the extracted pixel values;

a thinning module configured to thin the categorized pixels;

the processor further configured to convert the thinned categorized pixels; and

a correlator configured to cross correlate corresponding thinned pixels from consecutive frames.

15. The moving text detection device of 14 wherein the correlator is a phase plane correlator.

16. The moving text detection device of 13 wherein the processor is further configured to:

subtract a first pixel value from a second pixel value and obtaining a result; and

comparing the result to at least one threshold.

17. The moving text detection device of 14 wherein the thinning module is further configured to determine a first instance of similarly categorized pixels.

18. A computer readable storage medium comprising:

a first set of instructions adapted to provide a processor, wherein the processor is configured to implement a second set of instructions, the second set of instructions comprising;

an extraction code segment for extracting a sequence of pixels from consecutive frames; and

an identification code segment for identifying peak candidates derived from pixels that equal or exceed a threshold.

19. The computer readable storage medium of claim 18 wherein the second set of instructions further comprises:

a receiving code segment for receiving consecutive frames from a video stream.

20. The computer readable storage medium of claim 18 wherein the second set of instructions further comprises:

a categorization code segment for categorizing the pixels;

a thinning code segment for thinning the categorized pixels; and

a correlation code segment for correlating corresponding thinned pixels in the frames.

21. The computer readable medium of claim 18 wherein the first set of instructions or the second set of instructions are hardware description language (HDL) instructions.