Image compression apparatus, image compression program and image compression method

• US 20070248271A1
• Filed: 07/13/2006
• Published: 10/25/2007
• Est. Priority Date: 04/24/2006
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Noda, Tsugio, Sakai, Kenichiro
• Granted Patent: US 7,676,102 B2
• Time in Patent Office: Days
• US Class Current: 382/236
• CPC Class Codes: H04N 19/103 Selection of coding mode or of prediction mode
• CPC Class Codes: H04N 19/107 between spatial and temporal predictive coding, e.g. picture refresh
• CPC Class Codes: H04N 19/152 by measuring the fullness of the transmission buffer
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/182 the unit being a pixel
• CPC Class Codes: H04N 19/192 the adaptation method, adaptation tool or adaptation type being iterative or recursive
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/62 by frequency transforming in three dimensions H04N19/63 takes precedence
• CPC Class Codes: H04N 19/649 the transform being applied to non rectangular image segments

Abstract

Disclosed are an image compression apparatus, an image compression program, and an image compression method capable of increasing the compression ratio in lossless compression of a 3D image.

An image compression apparatus comprises: an image dividing section that divides the 3D image into a plurality of compression units each of which is a 3D area having a predetermined size; a pixel selection section that sequentially selects pixels in the compression unit as a target pixel in a predetermined order for each compression unit that has been obtained by the image dividing section and selects pixels located within a predetermined range near the target pixel as reference pixels; a prediction error calculation section that calculates a prediction value of the pixel value of the target pixel selected by the pixel selection section based on the pixel values of the reference pixels selected by the pixel selection section and calculates a difference between the prediction value and pixel value of the target pixel selected by the pixel selection section as a prediction error; and an entropy coding section that performs entropy coding of the prediction error calculated by the prediction error calculation section.

Claims

1. An image compression apparatus that performs lossless compression of a 3D image constituted by a plurality of frames, comprising:

an image dividing section that divides the 3D image into a plurality of compression units each of which is a 3D area having a predetermined size;

a pixel selection section that sequentially selects pixels in the compression unit as a target pixel in a predetermined order for each compression unit that has been obtained by the image dividing section and selects pixels located within a predetermined range near the target pixel as reference pixels;

a prediction error calculation section that calculates a prediction value of the pixel value of the target pixel selected by the pixel selection section based on the pixel values of the reference pixels selected by the pixel selection section and calculates a difference between the prediction value and pixel value of the target pixel selected by the pixel selection section as a prediction error; and

an entropy coding section that performs entropy coding of the prediction error calculated by the prediction error calculation section.

2. The image compression apparatus according to claim 1, wherein

the prediction error calculation section calculates the prediction value of the target pixel by summating values each obtained by multiplying the pixel value of each of the plurality of reference pixels by a weighting factor, the weighting factor being calculated such that the prediction error becomes minimum in each compression unit.

3. The image compression apparatus according to claim 2, further comprising a compressed data output section that outputs information concerning the 3D image, weighting factor calculated by the prediction error calculation section, and result of the entropy coding output by the entropy coding section as compressed data.

4. The image compression apparatus according to claim 3, further comprising a compression condition determination section that determines the number and position of the reference pixels such that the size of the compressed data output by the compressed data output section becomes minimum.

5. The image compression apparatus according to claim 4, wherein

the compression condition determination section determines the size of the compression unit such that the size of the compressed data output by the compressed data output section becomes minimum.

6. The image compression apparatus according to claim 3, further comprising a decompression section that calculates a prediction value of a target pixel in the same manner as the prediction error calculation section does based on the compressed data output by the compressed data output section, calculates a prediction error of the target pixel by decoding processing corresponding to the entropy coding performed by the entropy coding section, and restores the original pixel value of the target pixel based on the prediction value and prediction error of the target pixel.

7. The image compression apparatus according to claim 1, wherein

the image dividing section compares pixels values of the same positions in the frames that constitute the image to detect an area to be excluded from a compression target, and

the pixel selection section selects the target pixel and reference pixels from among pixels other than the detected area.

8. The image compression apparatus according to claim 1, wherein

the pixel selection section selects pixels located within a predetermined range near the target pixel in the same frame as that includes the target pixel and further selects pixels located within a predetermined range near the pixel at the same positions as the target pixel across frames within a predetermined range near the frame including the target pixel.

9. The image compression apparatus according to claim 1, wherein

the pixel selection section selects the reference pixels starting from a pixel nearest to the target pixel until the number of the reference pixels reaches a predetermined number.

10. The image compression apparatus according to claim 1, wherein

the 3D image is an image in which there is a correlation between adjacent frames.

11. An image compression program allowing a computer to execute lossless compression of a 3D image constituted by a plurality of frames, comprising:

an image dividing step that divides the 3D image into a plurality of compression units each of which is a 3D area having a predetermined size;

a pixel selection step that sequentially selects pixels in the compression unit as a target pixel in a predetermined order for each compression unit that has been obtained by the image dividing step and selects pixels located within a predetermined range near the target pixel as reference pixels;

a prediction error calculation step that calculates a prediction value of the pixel value of the target pixel selected by the pixel selection step based on the pixel values of the reference pixels selected by the pixel selection step and calculates a difference between the prediction value and pixel value of the target pixel selected by the pixel selection step as a prediction error; and

an entropy coding step that performs entropy coding of the prediction error calculated by the prediction error calculation step.

12. The image compression program according to claim 11, wherein

the prediction error calculation step calculates the prediction value of the target pixel by summating values each obtained by multiplying the pixel value of each of the plurality of reference pixels by a weighting factor, the weighting factor being calculated such that the prediction error becomes minimum in each compression unit.

13. The image compression program according to claim 12, further comprising, after the entropy coding step, a compressed data output step that outputs information concerning the 3D image, weighting factor calculated by the prediction error calculation step, and result of the entropy coding output by the entropy coding step as compressed data.

14. The image compression program according to claim 13, further comprising, before the image dividing step, a compression condition determination step that determines the number and position of the reference pixels such that the size of the compressed data output by the compressed data output step becomes minimum.

15. The image compression program according to claim 14, wherein

the compression condition determination step determines the size of the compression unit such that the size of the compressed data output by the compressed data output step becomes minimum.

16. The image compression program according to claim 13, further comprising a decompression step that calculates a prediction value of a target pixel in the same manner as the prediction error calculation step does based on the compressed data output by the compressed data output step, calculates a prediction error of the target pixel by decoding processing corresponding to the entropy coding performed by the entropy coding step, and restores the original pixel value of the target pixel based on the prediction value and prediction error of the target pixel.

17. The image compression program according to claim 11, wherein

the image dividing step compares pixel values of the same positions in the frames that constitute the image to detect an area to be excluded from a compression target, and

the pixel selection step selects the target pixel and reference pixels from among pixels other than the detected area.

18. The image compression program according to claim 11, wherein

the pixel selection step selects pixels located within a predetermined range near the target pixel in the same frame as that includes the target pixel and further selects pixels located within a predetermined range near the pixel at the same positions as the target pixel across frames within a predetermined range near the frame including the target pixel.

19. The image compression program according to claim 11, wherein

the pixel selection step selects the reference pixels starting from a pixel nearest to the target pixel until the number of the reference pixels reaches a predetermined number.

20. An image compression method that performs lossless compression of a 3D image constituted by a plurality of frames, comprising:

an image dividing step that divides the 3D image into a plurality of compression units each of which is a 3D area having a predetermined size;

a pixel selection step that sequentially selects pixels in the compression unit as a target pixel in a predetermined order for each compression unit that has been obtained by the image dividing step and selects pixels located within a predetermined range near the target pixel as reference pixels;

a prediction error calculation step that calculates a prediction value of the pixel value of the target pixel selected by the pixel selection step based on the pixel values of the reference pixels selected by the pixel selection step and calculates a difference between the prediction value and pixel value of the target pixel selected by the pixel selection step as a prediction error; and

an entropy coding step that performs entropy coding of the prediction error calculated by the prediction error calculation step.