METHOD AND SYSTEM FOR GENERATING A QUALITY PREDICTION TABLE FOR QUALITY-AWARE TRANSCODING OF DIGITAL IMAGES

• US 20090141992A1
• Filed: 06/30/2008
• Published: 06/04/2009
• Est. Priority Date: 12/03/2007
• Status: Active Grant

• Current Assignee: Ecole de technologie superieure, Vantrix Corporation
• Original Assignee: Ecole de technologie superieure
• Inventors: FRANCHE, Jean-Francois, COULOMBE, Stephane, PIGEON, Steven
• Granted Patent: US 8,295,624 B2
• Time in Patent Office: Days
• US Class Current: 382/235
• CPC Class Codes: G06T 9/004 Predictors, e.g. intraframe, interframe coding
• CPC Class Codes: H04N 19/115 Selection of the code volume for a coding unit prior to coding
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/149 by estimating the code amount by means of a model, e.g. mathematical model or statistical model
• CPC Class Codes: H04N 19/154 Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion use of rate-distortion criteria H04N19/147
• CPC Class Codes: H04N 19/192 the adaptation method, adaptation tool or adaptation type being iterative or recursive
• CPC Class Codes: H04N 19/40 using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream

Abstract

Several quality-aware transcoding systems and methods are described, in which the impact of both quality factor (QF) and scaling parameter choices on the quality of transcoded images are considered in combination. A basic transcoding system is enhanced by the addition of a quality prediction look-up table, and a method of generating such a table is also shown.

Claims

1. A system for generating a quality prediction table for predicting quality in image transcoding including:

a computer, comprising;

a transcoding module for transcoding an input image into an output image with a range of transcoder scaling factors and a range of output encoding quality factors;

a quality assessment module for determining a quality metric for each transcoding through a comparison of each output image with the corresponding input image;

a computer accessible storage medium available for storing the computed quality metrics in the form of a quality prediction table, having storage locations for computed quality metrics indexed by the respective transcoder scaling factor and output encoding quality factor used in each transcoding;

a computation module for creating the quality prediction table, creating the range of transcoder scaling factors and the range of the output encoding quality factors, obtaining the input images to be transcoded, and updating the computed quality metrics from each transcoding in the quality prediction table at the respective index location.

2. The system of claim 1, further including a computer accessible storage medium for storing a training set comprised of a number of the input images stored thereon.

3. The system of claim 1, further including an image feature extraction module for determining an input encoding quality factor of the input image, the encoding quality factor providing a further index into the quality prediction table.

4. The system of claim 1, wherein the quality prediction table is further indexed by viewing condition.

5. The system of claim 4, wherein the viewing condition is chosen to optimize an image quality experienced by a viewer of the output image.

6. The system of the claim 1, wherein the quality metrics is based on Peak Signal to Noise Ratio (PSNR) of the output image compared with the input image.

7. The system of the claim 1, wherein the quality metrics is based on Maximum Distance (MD) measure of the output image compared with the input image.

8. The system of the claim 1 wherein the transcoder scaling factor is incremented from 10% to 100% in steps of 10%.

9. The system of the claim 1 wherein the output encoding quality factor is incremented from 10 to 100 in steps of 10.

10. The system of the claim 1 where the input and output images are JPEG images.

11. A method for generating a quality prediction table for predicting quality in image transcoding, comprising steps of:

(a) obtaining an input image from a training set of images;

(b) transcoding the input image into an output image with a transcoder scaling factor from the range of transcoder scaling factors, and an output encoding quality factor from the range of output encoding quality factors;

(c) computing a quality metric for the transcoding through a comparison of the output image with the corresponding input image;

(d) updating a quality prediction table with the computed quality metrics;

(e) repeating the steps (b) to (e) for a range of transcoder scaling factors and a range of output encoding quality factors;

(f) repeating the steps (a) to (e) for all images in the training set.

12. The method of claim 11, wherein the step (d) comprises storing the computed quality metrics in the quality prediction table at a memory storage indexed by the respective transcoder scaling factor and output encoding quality factor used in the transcoding;

13. The method of claim 11, further comprising a step of determining an input encoding quality factor of the input image, and indexing the quality prediction table by the encoding quality factor.

14. The method of claim 11, further comprising indexing the quality prediction table by a viewing condition.

15. The method of claim 14, further comprising choosing the viewing condition to optimize an image quality experienced by a viewer of the output image.

16. The method of the claim 11, wherein the step (c) further includes determining the quality metrics based on a computation of Peak Signal to Noise Ratio (PSNR) of the output image compared with the input image.

17. The method of the claim 11, wherein the step (c) further includes determining the quality metrics based on a computation of Maximum Distance (MD) measure of the output image compared with the input image.

18. The method of the claim 11, wherein the step (e) includes incrementing the transcoder scaling factor from 10% to 100% in steps of 10%.

19. The method of the claim 11, wherein the step (e) includes incrementing the output encoding quality factor from 10 to 100 in steps of 10.

20. The method of the claim 11 where the input and output images are JPEG images.

21. The method of claim 11, further comprising a step of generating intermediate entries in the table by interpolating between the updated table entries.