SOFT DECISION MAKING PROCESSES FOR ANALYZING IMAGES

• US 20110044538A1
• Filed: 08/24/2009
• Published: 02/24/2011
• Est. Priority Date: 08/24/2009
• Status: Active Grant

• Current Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Original Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Inventors: Vanderhoff, Earl W.
• Granted Patent: US 9,230,173 B2
• Time in Patent Office: Days
• US Class Current: 382/165
• CPC Class Codes: G06V 20/46 Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames

Abstract

A device may calculate a normalized value for each of a number of pixels in a frame of a video stream, by obtaining a first color from one of the pixels and a second color, by obtaining color components of the first color and the second color, by, for each of the color components, determining a distance between the first color and the second color, and by adding the distances of the color components to obtain the normalized value. In addition, the device may compute an accumulation of the normalized values, compare the accumulation to a threshold to determine whether a first image that includes the pixels matches a second image that includes the second color, and display a result of determining whether the first image matches the second image via a graphical user interface (GUI).

Claims

1. A computer-readable medium including computer-executable instructions, the computer-executable instructions comprising:

instructions for obtaining a normalized value for each of a plurality of pixels, including instructions for;

obtaining a first color from one of the plurality of pixels,

extracting color components from the first color and a second color,

determining a distance for each of the color components between the first and second colors, and

combining the distances for the color components to obtain the normalized value; and

at least one of;

instructions for accumulating the normalized values to a generate a result, comparing the result to a threshold to determine whether a first image that includes the pixels matches a second image that includes the second color, and displaying a result of the comparison via a graphical user interface;

instructions for determining whether one of the pixels that includes the first color is to be copied to a whiteboard, and displaying the whiteboard via a graphical user interface;

or

instructions for determining whether a third image that includes the pixels is detected as a foreground image of a fourth image that includes the second color and displaying a result of the determination via a graphical user interface.

2. The computer-readable medium of claim 1, further comprising instructions for:

receiving a selection of an area that includes the first image.

3. The computer readable medium of claim 1, wherein the instructions for combining the distances include instructions for:

multiplying a sum of the distances by a coefficient; and

when a result of multiplying the sum by the coefficient exceeds a constant, setting the value to be equal to the constant.

4. The computer-readable medium of claim 1, wherein the instructions for determining a distance for each of the color components include instructions for at least one of:

computing a difference between a color component of the first color and a corresponding color component of the second color;

or

computing an absolute value of the difference.

5. The computer-readable medium of claim 1, wherein the instructions for combining the distances include instructions for:

adding the distances.

6. The computer-readable medium of claim 1, wherein the instructions for determining whether the third image is detected as a foreground image of the fourth image include at least one of:

instructions for determining how much of the third image matches a background image of the fourth image;

or

instructions for determining how much of the third image does not match the foreground image of the fourth image.

8. The computer-readable medium of claim 1, wherein the first image includes a portion of a frame of a video stream.

9. The computer-readable medium of claim 1, further comprising at least one of:

instructions for detecting a motion in a video stream that includes one of the second image or the foreground image of the fourth image;

instructions for recognizing images of an object in a video stream, the images of the object including one of the second image or the foreground image of the fourth image;

instructions for detecting an event in a video stream, the event being associated with the second image or the foreground image of the fourth image;

or

instructions for verifying a command that causes a device to output the second image or the foreground image of the fourth image.

10. The computer-readable medium of claim 1, wherein the second color includes at least one of:

a color selected by a user;

or

an average color of an image in the whiteboard.

11. The computer-readable medium of claim 1, wherein the color components include:

red, green, and blue.

7. The computer-readable medium of claim 7, wherein the fourth image includes a whiteboard image.

12. A method comprising:

calculating a normalized value for each of a plurality of pixels in a frame of a video stream, including;

obtaining a first color from one of the pixels and a second color,

obtaining color components of the first color and the second color,

for each of the color components, determining a distance between the first color and the second color, and

adding the distances of the color components to obtain the normalized value;

computing an accumulation of the normalized values;

comparing the accumulation to a threshold to determine whether a first image that includes the pixels matches a second image that includes the second color; and

displaying a result of determining whether the first image matches the second image via a graphical user interface (GUI).

13. The method of claim 12, wherein obtaining color components includes:

dividing a number representing the first color by a constant; and

obtaining a modulo of a result of the division.

14. The method of claim 12, wherein determining a distance includes:

calculating a difference between a component of the first color and a component of the second color.

15. The method of claim 12, wherein accumulating the normalized value includes:

a value obtained by multiplying the accumulation by a constant.

16. The method of claim 12, further comprising:

selecting the first image in the frame via the graphical user interface (GUI).

17. A device comprising:

a communication interface to receive a video stream from a remote device; and

a processor to;

obtain an image from the video stream,

obtain a reference color from a user via a graphical user interface,

for each pixel in the image, determine a distance between the pixel and the second color and normalize the distance,

add the normalized distances of the pixels to obtain a sum,

compare the sum to a threshold to determine whether the image matches a reference image or to determine whether the image is a foreground image of a reference image, and

display a result of comparing the sum to the threshold.

18. The device of claim 17, further comprising:

an infrared output port to issue commands to the remote device.

19. The device of claim 17, wherein the remote device includes:

a set-top box, a digital television that includes a plug-in cable card, or a computer.

20. The device of claim 17, wherein the reference image includes a white boarded image.