Adaptive image data compression

• US 6,724,817 B1
• Filed: 06/05/2000
• Issued: 04/20/2004
• Est. Priority Date: 06/05/2000
• Status: Expired due to Term

• Current Assignee: Dynamic Data Technologies LLC
• Original Assignee: Amphion Semiconductor Ltd. (Synaptics Incorporated)
• Inventors: Hu, Yi, Simpson, Thomas Albert
• Primary Examiner(s): DIEP, NHON THANH
• Application Number:
  US09/588,266
• Time in Patent Office: 1,415 Days
• Field of Search: H04/N.712, 375/240.07, 375/240.03, 375/240.04, 375/240.05, 375/240.06, 348/607
• US Class Current: 375/240.07
• CPC Class Codes: H04N 1/333 Mode signalling or mode changing; Handshaking therefor
• CPC Class Codes: H04N 19/15 by monitoring actual compressed data size at the memory before deciding storage at the transmission buffer
• 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 19/18 the unit being a set of transform coefficients
• CPC Class Codes: H04N 19/196 being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters processing of motion vectors H04N19/513
• CPC Class Codes: H04N 19/60 using transform coding

Abstract

The invention relates to a method and apparatus for compressing an image data signal. The apparatus includes a compression control module arranged to eliminate frequency components produced from the image data signal which have a magnitude less than a threshold value. The apparatus compares the volume of the compressed image data signal with a target volume and selects to increase or decrease the applied compression level depending on the outcome of the comparison. The compression level is increased by increasing the threshold value and deacreased by decreasing the threshold value. The compression level, or ratio, is thus adaptable in real-time in order to meet a target volume for the compressed image. This is particularly advantageous in applications where, for example, the compressed image signal is to be transmitted over a fixed bandwidth link. The method and apparatus of the invention are particularly suitable for use in the implementation of the JPEG basline image compression standard.

Claims

1. An image compression apparatus for compressing an image data signal comprising one or more image data blocks, the apparatus comprising a transform module arranged to transform the or each image data block into an array of frequency coefficients;

a quantization module arranged to quantize each of said frequency coefficients according to a respective quantization coefficient;

a compression control module arranged to eliminate each frequency coefficient having a magnitude less than a threshold value;

an entropy encoding module arranged to generate a coded data element for each non-zero and non-eliminated frequency coefficient to produce a compressed data signal, wherein the image data signal is processed in image segments, each image segment comprising at least one image data block, and a respective target volume is set for each image segment, the apparatus being arranged to measure the volume of the compressed data signal produced for each image segment, to calculate the difference between the respective target volume and respective measured volume for each image segment, and to sum the calculated differences, the compression control module being arranged to adapt the level of compression applied by the apparatus depending on the value of said summed calculated differences.

2. An apparatus as claimed in claim 1, wherein the compression control module is arranged to increase the level of compression by increasing the threshold value, and to decrease the level of compression by decreasing the threshold value.

3. An apparatus as claimed in claim 1, wherein the compression control module is arranged to adapt selectively the applied compression level after an image segment is processed, the selection further depending on the value of the calculated difference for said image segment.

4. An apparatus as claimed in claim 3, wherein the compression control module selects to increase the applied compression level if, after processing an image segment, the sum of the calculated differences is negative and the individual calculated difference for said image segment is also negative, and selects to decrease the applied compression level if the sum of the calculated differences is positive and the individual calculated difference for said image segment is also positive.

5. An apparatus as claimed in claim 1, wherein each applicable compression level is associated with more than one threshold value, the compression control apparatus being arranged to apply a respective threshold value to one or more frequency coefficients depending on the spectral significance of said one or more frequency coefficients in the image data signal.

6. An apparatus according to claim 1, wherein the compression controller is arranged between the quantization module and the entropy encoder, and to operate on the quantized frequency coefficients.

7. An apparatus according to claim 1, wherein the apparatus is arranged to implement sequential JPEG encoding, the entropy encoding module including a run length coding module and a Huffman encoder, the compression controller being arranged to supply the non-zero quantized frequency coefficients to the Huffman encoder via the run length coding module, the Huffman encoder being arranged to produce a respective coded data element for each non-zero quantized frequency coefficient, to measure the volume of each produced coded data element and to communicate the measured volume to the compression control module.

8. An apparatus as claimed in claim 7, wherein the Huffman encoder is arranged to sum said measured volumes for each image data block, to signal to the compression control module when each image data block is processed and to communicate to the compression control module the respective sum of the measured volumes for each image data block.

9. In an image compression apparatus for compressing an image data signal comprising one or more image data blocks, the apparatus comprising a transform module arranged to transform the or each image data block into an array of frequency coefficients;

a quantization module arranged to quantize each of said frequency coefficients according to a respective quantization coefficient;

a compression control module arranged to eliminate each frequency coefficient having a magnitude less than a threshold value; and

an entropy encoding module arranged to generate a coded data element for each non-zero frequency coefficient to produce a compressed data signal, wherein the image data signal is processed in image segments, each image segment comprising at least one image data block, and a respective target volume is set for each image segment, a method of adapting the level of compression applied by the apparatus, the method comprising;

measuring the volume of the compressed data signal produced for each image segment;

calculating the difference between the respective target volume and respective measured volume for each image segment;

summing the calculated differences; and

adapting the level of compression applied by the apparatus depending on the value of said summed calculated differences.

10. The image compression apparatus as claimed in claim 9, wherein the applied compression level is increased by increasing the threshold value, and decreased by decreasing the threshold value.

11. The image compression apparatus as claimed in claim 10, wherein the applied compression level is selectively adapted after an image segment is processed, the selection further depending on the value of the calculated difference for said image segment.

12. The image compression apparatus as claimed in claim 11, further including increasing the applied compression level if, after processing an image segment, the sum of the calculated differences is negative and the individual calculated difference for said image segment is also negative;

or decreasing the applied compression level if the sum of the calculated differences is positive and the individual calculated difference for said image segment is also positive.

13. The image compression apparatus as claimed in claim 9, wherein each applicable compression level is associated with more than one threshold value, the method further including applying a respective threshold value to one or more frequency coefficients depending on the spectral significance of said one or more frequency coefficients in the image data signal.

14. The image compression apparatus as claimed in claim 9, wherein the compression controller is arranged to operate on the quantized frequency coefficients.

15. The image compression apparatus as claimed in claim 9, wherein the apparatus is arranged to implement sequential JPEG encoding, the entropy encoding module including a run length coding module and a Huffman encoder, the method further including supplying the non-zero quantized frequency coefficients to the Huffman encoder via the run length coding module, producing a respective coded data element for each non-zero quantized frequency coefficient;

measuring the volume of each produced coded data element; and

communicating the measured volume to the compression control module.

16. The image compression apparatus as claimed in claim 15, the method further including summing the measured volumes for each image data block;

signalling to the compression control module when each image data block is processed; and

communicating to the compression control module the respective sum of the measured volumes for each image data block.