Frame averaging for use in processing video data

• US 6,005,638 A
• Filed: 03/04/1996
• Issued: 12/21/1999
• Est. Priority Date: 03/04/1996
• Status: Expired due to Term

• Current Assignee: Xylon LLC (Intellectual Ventures LLC)
• Original Assignee: Axcess, Inc.
• Inventors: Blair, Bruce A., Parvini, Morteza
• Primary Examiner(s): LEE, MICHAEL
• Application Number:
  US08/610,618
• Time in Patent Office: 1,387 Days
• Field of Search: 348/607, 348/608, 348/613, 348/618, 348/619, 348/620, 348/621, 348/700, 348/701, 348/595, 348/597, 348/459, 348/616, 348/617, 348/578, 348/579
• US Class Current: 348/607
• CPC Class Codes: H04N 5/21 Circuitry for suppressing or minimising disturbance, e.g. moiré or halo

Abstract

The invention enables processing of a time-varying video image in a manner that blurs (or "smears") the time-varying portion or portions of the video image as successive frames of video data are processed. Such smearing is particularly useful in reducing the choppiness and jerkiness that degrades the viewability of the time-varying portions of a displayed video image when the frame rate is low. The invention blends corresponding video data components from a current video frame and a previous video frame to achieve the smearing effect. The invention enables the degree to which the video data from each of the frames affects the processed video data to be varied. The exact blending can be based upon, for example, the magnitude of the frame rate. Further, in processing each video frame, each of a set of related video data components can be processed serially to enable the invention to be implemented more compactly and inexpensively. The invention can be implemented in a video data capture part of a video system, a video data display part of a video system, or in both a video data capture and video data display parts. In addition to producing smearing, the invention reduces noise in the video image.

Claims

1. A system for processing a plurality of sequentially occurring frames of time-varying video data in order to blend discontinuities in the images in the frames together, the video data including a plurality of sets of related video data components, the system comprising:

means for receiving successive sets of input signals in parallel and serially outputting the input signals of each set as current frame signals, each of the successive sets of input signals representing the magnitudes of a set of related video data components of a frame of unprocessed video data;

means for storing a plurality of values representing magnitudes of sets of related video data components for a previous frame of video data;

a current image content gain device for receiving each of the current frame signals output by the means for receiving and for outputting corresponding amplified current frame signals;

a prior image content gain device for receiving previous frame signals representing the magnitudes of the video data components for the previous frame of video data, the previous frame signals being generated from the values in a memory, and for outputting corresponding amplified previous frame signals, the current and prior image content gain devices being controlled so that the amplified current frame signals and amplified previous frame signals for corresponding video data components are output at the same time;

a blender for receiving corresponding amplified current frame signals and amplified previous frame signals and for outputting output signals representing the sum of the corresponding amplified current frame signals and amplified previous frame signals to blend discontinuities in the current frame signals and the previous frame signals together, the output signals being used to update the values in the memory for the corresponding video data components of the previous frame of video data; and

means for receiving the output signals serially from the adder and outputting in parallel sets of output signals corresponding to sets of input signals, each set of output signals representing the magnitudes of a set of related video data components of a current frame of processed video data.

2. A system as in claim 1, wherein the means for receiving and serially outputting, the current image content gain device, the prior image content gain device, the adder, and the means for receiving serially and outputting in parallel are each implemented in a field programmable gate array.

3. A system as in claim 1, wherein the frame rate is less than or equal to approximately 24 frames per second.

4. A system as in claim 1, wherein:

the means for receiving and serially outputting is a multiplexer; and

the means for receiving serially and outputting in parallel is a demultiplexer.

5. A system for processing a plurality of sequentially occurring frames of time-varying video data to blend discontinuities in the sequentially occurring frames together, the video data including a plurality of video data components, the system comprising:

means for acquiring a plurality of frames of input video data at a frame rate of less than or equal to approximately 24 frames per second;

a frame blending device for generating a plurality of signals representing weighted averages of the values of corresponding video data components from a currently acquired frame of input video data and a previous frame of input video data in order to blend discontinuities between the currently acquired frame of input video data and the previous frame of input video data together;

means for transmitting the signals to a system for displaying the video image;

means for receiving the signals transmitted by the means for transmitting;

a second frame averaging device for generating a second plurality of signals representing weighted averages of the values of corresponding video data components from a previously displayed frame of video data and a current frame of video data represented by the signals received by the means for receiving; and

a display device for displaying output video data, the output video data being represented by the second plurality of signals.

6. A system as in claim 5, further comprising a video data compressor for compressing the plurality of signals before the signals are transmitted by the means for transmitting.

7. A system as in claim 6, further comprising:

means for receiving the compressed signals transmitted by the means for transmitting;

a video data decompressor for decompressing the compressed signals;

a second frame averaging device for generating a second plurality of signals representing weighted averages of the values of corresponding video data components from a previously displayed frame of video data and a current frame of video data represented by the decompressed signals; and

a display device for displaying output video data, the output video data being represented by the second plurality of signals.

8. A system for processing a plurality of sequentially occurring frames of time-varying video data to blend discontinuities in the sequentially occurring frames together, comprising:

means for acquiring a current frame signal representing the value of a video data component from a current frame of unprocessed video data;

means for generating a previous frame signal representing the value of a video data component from a previous frame of video data that corresponds to the video data component from the current frame of unprocessed video data;

frame blending means for receiving the current frame signal and the previous frame signal, and outputting an output signal representing processed video data for the current frame, the output signal representing a weighted average of the values of the corresponding video data components from the current frame of unprocessed video data and the previous frame of video data to blend discontinuities between the current frame of unprocessed video data and the previous frame of video data together;

means for varying the weights associated with the current frame of unprocessed video data and the previous frame of video data, operably coupled to the frame averaging means, to enable variation of the relative degree of influence on the output signal of the video data components from the current frame of unprocessed video data and the previous frame of video data; and

wherein the frame averaging means and means for varying further comprise a mixture selector for receiving the current video signal, the previous video signal and a frame weight selection signal, and outputting a first intermediate signal, the frame weight selection signal representing a particular set of weights associated with the current frame of unprocessed video data and the previous frame of video data, the first intermediate signal being equivalent to either the current or previous video signal according to the set of weights represented by the frame weight selection signal, a first adder for receiving the current video signal and the first intermediate signal, and outputting a second intermediate signal that is the sum of the current video signal and the first intermediate signal, and a second adder for receiving the second intermediate signal and the previous video signal, and outputting the sum of the second intermediate signal and the previous video signal as the weighted average signal.

9. A system as in claim 8, wherein the frame averaging means and means for varying further comprise a digital video mixer.

10. A system as in claim 8, wherein the weights associated with the current frame of unprocessed video data and previous frame of video data are dependent upon the magnitude of the frame rate.

11. A system as in claim 8, further comprising means for determining the frame rate of the video data.

12. A system as in claim 8, wherein:

for frame rates less than or equal to approximately five frames per second, the means for varying associates a weight of 0.75 with the previous frame of video data and a weight of 0.25 with the current frame of unprocessed video data; and

for frame rates greater than approximately five frames per second and less than or equal to approximately 24 frames per second, the means for varying associates a weight of 0.5 with the previous frame of video data and a weight of 0.5 with the current frame of unprocessed video data.

13. A method for processing a plurality of sequentially occurring frames of time-varying video data to blend discontinuities in the sequentially occurring frames together, comprising the steps of:

determining the value of a video data component from a current frame of unprocessed video data;

determining the value of a corresponding video data component from a previous frame of video data;

forming a weighted average of the values of the corresponding video data components from the current frame of unprocessed video data and the previous frame of video data to blend discontinuities between the current frame of unprocessed video data and the previous frame of video data together, wherein the weights associated with the current frame of unprocessed video data and previous frame of video data can be varied according to a predefined instruction to enable variation of the relative degree of influence on the weighted average of the video data components from the current frame of unprocessed video data and the previous frame of video data; and

wherein the step of forming further comprises inputting the value of the video data component from the current frame of unprocessed video data, inputting the value of the video data component from the previous frame of video data, inputting a frame weight selection representing a particular set of weights associated with the current frame of unprocessed video data and the previous frame of video data, selecting either the value of the video data component from the current frame of unprocessed video data or the value of the video data component from the previous frame of video data, depending upon the frame weight selection, as a value of a first intermediate weighted data component, adding the value of the video data component from the current frame of unprocessed video data to the value of the first intermediate weighted data component to produce a value of a second intermediate weighted data component, and adding the value of the second intermediate weighted data component and the value of the video data component from the previous frame of video data to produce the weighted average.

14. A method as in claim 13, wherein the weights associated with the current frame of unprocessed video data and the previous frame of video data are dependent upon the magnitude of the frame rate.

15. A method as in claim 14, further comprising the step of determining the frame rate of the video data.

16. A method as in claim 14, wherein the step of inputting a frame weight selection further comprises the steps of:

for frame rates less than or equal to approximately five frames per second, associating a weight of 0.75 with the previous frame of video data and a weight of 0.25 with the current frame of unprocessed video data; and

for frame rates greater than approximately five frames per second and less than or equal to approximately 24 frames per second, associating a weight of 0.5 with the previous frame of video data and a weight of 0.5 with the current frame of unprocessed video data.

17. A system for processing a plurality of sequentially occurring frames of time-varying video data in order to blend discontinuities in the images in the frames together, the video data including a plurality of sets of related video data components, the system comprising:

a multiplexer that receives successive sets of input signals in parallel and serially outputs the input signals of each set as current frame signals, each of the successive sets of input signals representing the magnitudes of a set of related video data components of a frame of unprocessed video data;

means for storing a plurality of values representing magnitudes of sets of related video data components for a previous frame of video data;

a current image content gain device for receiving each of the current frame signals output by the means for receiving and for outputting corresponding amplified current frame signals;

a prior image content gain device for receiving previous frame signals representing the magnitudes of the video data components for the previous frame of video data, the previous frame signals being generated from the values in a memory, and for outputting corresponding amplified previous frame signals, the current and prior image content gain devices being controlled so that the amplified current frame signals and amplified previous frame signals for corresponding video data components are output at the same time;

an adder for receiving corresponding amplified current frame signals and amplified previous frame signals and for outputting output signals representing the sum of the corresponding amplified current frame signals and amplified previous frame signals to blend discontinuities in the current frame signals and the previous frame signals together, the output signals being used to update the values in the memory for the corresponding video data components of the previous frame of video data; and

a demultiplexer that receives the output signals serially from the adder and outputs in parallel sets of output signals corresponding to sets of input signals, each set of output signals representing the magnitudes of a set of related video data components of a current frame of processed video data.