Perception-based audio visual synchronization measurement system

• US 5,596,364 A
• Filed: 03/30/1995
• Issued: 01/21/1997
• Est. Priority Date: 10/06/1993
• Status: Expired due to Fees

• Current Assignee: United States Secretary of Commerce
• Original Assignee: THE GOVERNMENT OF THE UNITED STATES OF AMERICA, AS REPRESENTED BY THE SECRETARY OF COMMERCE
• Inventors: Wolf, Stephen, Voran, Stephen, Jones, Coleen T., Kubichek, Robert F., Webster, Arthur A. III, Pinson, Margaret H.
• Primary Examiner(s): Peng, John K.
• Assistant Examiner(s): GRANT, CHRISTOPHER C
• Application Number:
  US08/413,362
• Time in Patent Office: 663 Days
• Field of Search: 348/192, 348/193, 348/194, 348/189, 348/180, 348/15, 348/16, 348/17, 348/18, 348/19, 358/10, 358/139
• US Class Current: 348/192
• CPC Class Codes: H04N 17/00 Diagnosis, testing or measuring for television systems or their details
• CPC Class Codes: H04N 5/04 Synchronising for television systems using pulse code modulation H04N7/56
• CPC Class Codes: H04N 5/60 for the sound signals

Abstract

A method for objectively and non-intrusively measuring the audio-visual quality of a destination audio-video signal that generates measurement parameters that are indicative of human image quality perceptions. Subjective human test panel results are generated for a variety of audio-video test samples and objective test results are also generated by the apparatus of the invention for the variety of audio-video test samples. Accordingly, when the apparatus extracts test frames from the actual source and destination audio-video signals and compares them, audio-visual quality parameters are output by the apparatus that are indicative of the audio-visual quality of the destination audio-visual signal based upon the source audio-visual signal.

Claims

1. A quality measurement system for measuring an audio-visual quality of a destination audio-visual signal, comprising:

source audio feature extraction means for extracting source audio features from a source audio-visual signal corresponding to the destination audio-visual signal;

source video feature extraction means for extracting source video features from a source audio-visual signal corresponding to the destination audio-visual signal;

destination audio feature extraction means for extracting destination audio features from the destination audio-visual signal;

destination video feature extraction means for extracting destination video features from the destination audio-visual signal;

source time referencing means for providing a source time reference for each one of the source audio features and for each one of the source video features;

destination time referencing means for providing a destination time reference for each one of the destination audio features and for each one of the destination video features;

communication circuit means for communicating the source audio features, the source video features, the source time references, the destination audio features, the destination video features, and the destination time references;

audio quality parameter generation means for generating audio quality parameters, indicative of the audio quality of the destination audio-visual signal based on the source audio-visual signal, from at least one source and at least one destination audio feature;

video quality parameter generation means for generating video quality parameters, indicative of the video quality of the destination audio-visual signal based on the source audio-visual signal, from at least one source and at least one destination video feature; and

audio-visual quality parameter generation means for generating audio-visual quality parameters, indicative of the audio-visual quality of the destination audio-visual signal based on the source audio-visual signal, from at least one audio quality parameter and at least one video quality parameter.

2. The quality measurement system of claim 1, wherein the audio-visual quality parameters are indicative of human audio-visual quality perception.

3. The quality measurement system of claim 1, wherein said audio quality parameter generation means generates an audio delay parameter based on at least the extracted source and destination audio features.

4. The quality measurement system of claim 3, wherein said audio quality parameter generation means generates the audio delay parameter further based on a communication circuit delay.

5. The quality measurement system of claim 3, wherein said audio quality parameter generation means generates the audio delay parameter further based on source and destination time references.

6. The quality measurement system of claim 1, wherein said audio-visual quality parameter generation means generates an audio-visual synchronization parameter based on at least the generated audio-visual quality parameters.

7. The quality measurement system of claim 1, wherein the audio-visual quality parameters include an audio-visual synchronization parameter indicative of the difference between time delays of the audio and video components of the destination audio-visual signal.

8. An audio-visual quality measurement system for measuring a change quality between a source audio-visual signal and a corresponding destination audio-visual signal, the audio-visual quality measurement system comprising:

first audio feature extraction means for extracting a first plurality of audio features from the source audio-visual signal based on at least one objective audio measurement criteria;

second audio feature extraction means for extracting a second plurality of audio features from the destination audio-visual signal based on the at least one objective audio measurement criteria;

audio alignment means coupled to the first and the second audio feature extraction means for performing time alignment between corresponding audio features of the first and the second plurality of audio features;

audio quality parameter generation means for combining time aligned corresponding features of the first and the second plurality of audio features to generate at least one audio quality indication;

first video feature extraction means for extracting a first plurality of video features from the source audio-visual signal based on at least one objective video measurement criteria;

second video feature extraction means for extracting a second plurality of video features from the destination audio-visual signal based on the at least one objective video measurement criteria;

video alignment means coupled to the first and the second video feature extraction means for performing time alignment between corresponding video features of the first and the second plurality of video features;

video quality parameter generation means for combining time aligned corresponding features of the first and the second plurality of video features to generate at least one video quality indication; and

audio-visual combining means for combining the time aligned corresponding audio and video quality parameters of the first and the second plurality of audio and video features according to a plurality of learned parameters, the plurality of learned parameters having been determined from a joint statistical analysis of a plurality of subjective viewer judgments.

9. The system of claim 8, wherein the first audio feature extraction means and the second audio feature extraction means each comprise:

low pass filtering means for generating a plurality of low pass filtered audio samples, each low pass filtered audio sample corresponding to a respective audio sample of the plurality of low pass filtered audio samples; and

sub-sampling means for determining a plurality of sub-sampled audio features, each sub-sampled audio feature corresponding to a respective one of the plurality of low pass filtered audio samples.

10. The system of claim 8, wherein the first video feature extraction means and the second video feature extraction means each comprise:

absolute frame difference filtering means for generating a plurality of difference filtered video frames corresponding to a respective video frame of the plurality of video frames and including a respective plurality of difference filtered pixels; and

temporal statistics generating means for determining a plurality of temporal statistic features, each temporal statistic feature corresponding to a statistic of the difference filtered pixels of a respective one of the plurality of difference filtered pixels.

11. A quality measurement system for measuring an audio-visual quality of a destination audio-visual signal, comprising:

source audio feature extraction means for extracting source audio features from a source electromagnetic audio-visual signal corresponding to the destination electromagnetic audio-visual signal;

destination audio feature extraction means for extracting destination audio features from the destination electromagnetic audio-visual signal;

source time referencing means for providing a source time reference for each one of the source audio features;

destination time referencing means for providing a destination time reference for each one of the destination audio features;

communication circuit means for communicating the source audio features, the source time references, the destination audio features, and the destination time references; and

audio quality parameter generation means for generating audio quality parameters, indicative of the audio quality of the destination electromagnetic audio-visual signal based on the source electromagnetic audio-visual signal, from at least one source and at least one destination audio feature.

12. The system of claim 11, further comprising:

source electromagnetic video feature extraction means for extracting source video features from a source electromagnetic audio-visual signal corresponding to the destination electromagnetic audio-visual signal;

destination video feature extraction means for extracting destination video features from the destination audio-visual signal, wherein said source time referencing means provides a source time reference for each one of the source video features, said destination time referencing means provides a destination time reference for each one of the destination video features, and said communication circuit means communicates source video features, source time references, destination video features, and destination time references;

video quality parameter generation means for generating video quality parameters, indicative of the video quality of the destination audio-visual signal based on the source audio-visual signal, from at least one source and at least one destination video feature;

audio-visual quality parameter generation means for generating audio-visual quality parameters, indicative of the audio-visual quality of the destination audio-visual signal based on the source audio-visual signal, from at least one audio quality parameter and at least one video quality parameter.