Image sequence compression and decompression

• US 5,841,473 A
• Filed: 10/31/1996
• Issued: 11/24/1998
• Est. Priority Date: 07/26/1996
• Status: Expired due to Fees

• Current Assignee: Foto-Wear, Inc.
• Original Assignee: Software for Image Compression N.V.
• Inventors: Chui, Charles K., Wang, Jianrong
• Primary Examiner(s): Britton, Howard W.
• Application Number:
  US08/732,491
• Time in Patent Office: 754 Days
• Field of Search: 348/390, 348/403, 348/421, 348/424, 348/425, 348/397, 348/398, 382/232, 382/248
• US Class Current: 348/390.1
• CPC Class Codes: H04N 19/115 Selection of the code volume for a coding unit prior to coding
• CPC Class Codes: H04N 19/146 Data rate or code amount at the encoder output
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/63 using sub-band based transform, e.g. wavelets
• CPC Class Codes: H04N 19/635 characterised by filter definition or implementation details

Abstract

A method and system for compressing and decompressing a sequence of image frames is disclosed. According to the disclosed embodiments of the invention, the compression process includes an intraframe decomposition process (24, 24'"'"') performed upon each frame in both the row and column directions, including an m^th order averaging and differencing operation (56, 62). Each frame is then combined with others in an adjacent pair or group of four by way of an interframe decomposition (28). Quantization (30) is then performed, preferably by division of each coefficient by a quantizer (ρ) associated with the one of the frequency component subbands containing the coefficient. Reconstruction of the compressed frames is accomplished by performing the inverse of the compression process. According to the disclosed embodiments, either the compression or decompression processes and systems performs Cholesky filtering (60, 66; 170, 172), preferably in a recursive manner by way of forward and backward filtering. In this way, one or the other of the compression system (CS) or decompression system (DS) is relieved from a significant amount of computing burden.

Claims

1. A method of compressing and decompressing a sequence of images, comprising the steps of:

compressing the sequence of images into a compressed sequence, by performing the steps of;

digitizing each image in the sequence into an image frame comprising an array of intensity coefficients arranged in first and second directions;

performing an intraframe decomposition of each image frame by the steps of;

performing averaging and differencing of the image frame in a first direction to generate average and difference coefficient sequences;

downsampling the average and difference coefficient sequences;

performing averaging and differencing of the image frame in a second direction; and

downsampling sequences from the step of performing averaging and differencing of the image frame in the second direction, to generate a decomposed image frame;

performing an interframe decomposition of groups of decomposed image frames to generate a compressed sequence of image frames;

decompressing the compressed sequence of images into a reconstructed sequence, by performing the steps of;

performing an interframe reconstruction of the groups of decomposed image frames; and

then performing an intraframe reconstruction of each image frame by the steps of;

upsampling the average and difference coefficient sequences;

performing average and difference combining of the upsampled sequences in the second direction;

upsampling sequences resulting from the step of performing average and difference combining; and

performing averaging and differencing of the image frame in the first direction, to generate a reconstructed image frame;

wherein one of the intraframe decomposition and intraframe reconstruction steps further comprises;

performing Cholesky filtering upon coefficient sequences corresponding to the image frame.

2. The method of claim 1, wherein the steps of performing averaging and differencing, of performing average and difference combining, and of performing Cholesky filtering, are performed according to an order m, where m is greater than one.

3. The method of claim 1, wherein the step of performing interframe decomposition comprises:

combining first and second decomposed image frames with one another to generate a frame pair comprising average frame and a difference frame; and

repeating the combining step for each pair of decomposed image frames in the sequence;

and wherein the step of performing interframe reconstruction comprises;

operating upon the average frame and difference frame in each frame pair to recover first and second image frames therefrom.

4. The method of claim 1, wherein the step of performing interframe decomposition comprises:

combining a group of four decomposed image frames with one another to generate a group of four transform frames; and

repeating the combining step for each group of four decomposed image frames in the sequence;

and wherein the step of performing interframe reconstruction comprises;

operating upon each group of four transform frames to recover a group of four image frames therefrom.

5. The method of claim 1, further comprising:

displaying a motion picture comprised of a sequence of reconstructed image frames.

6. The method of claim 1, wherein the step of performing an intraframe decomposition is performed a plurality of times upon each frame.

7. The method of claim 6, wherein the step of performing an intraframe reconstruction is performed, for each frame, the same number of times as the step of performing an intraframe decomposition.

8. The method of claim 1, wherein the step of performing an intraframe decomposition of each image frame further comprises:

performing Cholesky filtering upon the average and difference coefficient sequences; and

performing Cholesky filtering upon the results of the step of performing averaging and differencing in a second direction.

9. The method of claim 8, wherein each of the Cholesky filtering steps comprises:

performing a backward differencing; and

performing a forward differencing.

10. The method of claim 8, wherein the step of performing Cholesky filtering upon the average and difference coefficient sequences is performed after the first downsampling step;

and wherein the step of performing Cholesky filtering upon the results of the step of performing averaging and differencing in a second direction is performed after the second downsampling step.

11. The method of claim 1, wherein the step of performing an intraframe reconstruction of each image frame further comprises:

performing Cholesky filtering upon components of each image frame, before the step of performing average and difference combining in the second direction; and

performing Cholesky filtering upon the results of the step of performing average and difference combining in the second direction, before the step of performing average and difference combining in the first direction.

12. The method of claim 11, wherein each of the Cholesky filtering steps comprises:

performing a backward differencing; and

performing a forward differencing.

13. The method of claim 11, wherein the step of performing Cholesky filtering upon components of each image frame is performed before the first upsampling step;

and wherein the step of performing Cholesky filtering upon the results of the step of performing average and difference combining in the second direction is performed before the second upsampling step.

14. The method of claim 1, wherein the compressing step further comprises:

after the step of performing an interframe decomposition, quantizing each of the coefficients in the compressed sequence of image frames;

and wherein the decompressing step further comprises;

before the step of performing an interframe reconstruction, dequantizing the compressed sequence of images.

15. The method of claim 14, wherein the quantizing step comprises:

dividing each coefficient in the compressed sequence of image frames by a quantizer;

and wherein the dequantizing step comprises;

multiplying each coefficient in the compressed sequence of images by the quantizer.

16. The method of claim 15, wherein each image frame in the compressed sequence of image frames corresponds to a frequency component subband, such that each of the image frames corresponds contains a plurality of subbands;

and wherein the quantizer used in the quantizing and dequantizing steps varies among the plurality of subbands.