Estimated spectrum adaptive postfilter and the iterative prepost filtering algirighms

• US 6,760,487 B1
• Filed: 04/14/2000
• Issued: 07/06/2004
• Est. Priority Date: 04/22/1999
• Status: Expired due to Term

• Current Assignee: United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration
• Original Assignee: United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration
• Inventors: Linares, Irving
• Primary Examiner(s): Dastouri, Mehrdad
• Assistant Examiner(s): Miller, Martin
• Application Number:
  US09/550,254
• Time in Patent Office: 1,544 Days
• Field of Search: 382/261, 382/263, 382/266, 382/275, 341/67, 341/87, 341/94, 348/394, 348/610, 348/618, 375/240.02, 375/240.03, 375/240.24, 378/4, 704/200.1, 708/203
• US Class Current: 382/275
• CPC Class Codes: H04N 19/117 Filters, e.g. for pre-processing or post-processing sub-band filter banks H04N19/635
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties

Abstract

The invention presents The Estimated Spectrum Adaptive Postfilter (ESAP) and the Iterative Prepost Filter (IPF) algorithms. These algorithms model a number of image-adaptive post-filtering and pre-post filtering methods. They are designed to minimize Discrete Cosine Transform (DCT) blocking distortion caused when images are highly compressed with the Joint Photographic Expert Group (JPEG) standard. The ESAP and the IPF techniques of the present invention minimize the mean square error (MSE) to improve the objective and subjective quality of low-bit-rate JPEG gray-scale images while simultaneously enhancing perceptual visual quality with respect to baseline JPEG images.

Claims

1. A method for reducing image blocking artifacts resulting from high JPEG image compression comprising the steps of:

pre-emphasizing an original image file x with a filter P (ω

,G), to form a pre-emphasis signal X_e;

wherein ω

is the HPF cutoff frequency corresponding to the separable DCT bandwidth ƒ and

where (ω

≈

ƒ

) and G is the pre-emphasis gain;

compressing the dynamic range of said pre-emphasis signal X_e into an 8-BPP range of 0-255 using a dynamic range compression (DRC) function to form a range-compressed signal X_r;

using said range-compressed signal X_r to compute a MMSE optimally quantized Q-table using a Lagrange multiplier minimization function adaptQw( );

using parameters compiled in said Q-table;

compressing said range-compressed signal X_r image file x using the JPEG standard IJG cjpeg( ) function to form a compressed JPEG file X_i(k);

de-compressing said JPEG file X_i(k) to form a de-compressed image file x{circumflex over ( )}_r;

expanding said de-compressed image file using dynamic range expansion (DRE) to form an image file x{circumflex over ( )}_e;

de-emphasizing said image file x{circumflex over ( )}_e using a filter D(ω

, G) to create a x{circumflex over ( )} image;

post-filtering said x{circumflex over ( )} image into an x{tilde over ( )} image, and thereafter;

using an ESAP process to obtain further block smoothing of said x-image, wherein ESAP is Estimated Spectrum Adaptive Post-Filter.

2. The method of claim 1, wherein the ESAP process comprises the steps of:

providing a decompressed image file;

providing dequantized DCT coefficients;

estimating 2-D pixel-adaptive bandwidths directly from said dequantized DCT coefficients; and

using said dequantized DCT coefficients to control a 2-D spatially-adaptive non-linear image post-filter.

3. The method of claim 2, further comprising the step of:

using an edge detector to control a 2-D spatially-adaptive non-linear image post-filter.

4. The method of claim 2, wherein the edge detector is Canny edge detector.

5. A method for reducing image blocking artifacts resulting from high JPEG image compression comprising the steps of:

pre-emphasizing an original image file x with a filter P(ω

,G), to form a pre-emphasis signal X_e;

wherein ω

is the HPF cutoff frequency corresponding to the separable DCT bandwidth ƒ and

where (ω

≈

ƒ

) and G is the pre-emphasis gain;

compressing the dynamic range of said pre-emphasis signal X_e into an 8-EPP range of 0-255 using a dynamic range compression (DRC) function to form a range-compressed signal X_r;

using said range-compressed signal X_r to compute a MMSE optimally quantized Q-table using a Lagrange multiplier minimization function adaptQw( );

using parameters compiled in said Q-table;

compressing said range-compressed signal X_r image file x using the JPEG standard IJG cjpeg( ) function to form a compressed JPEG file X_i(k);

de-compressing said JPEG file X_i(k) to form a de-compressed image tile x{circumflex over ( )}_r;

expanding said de-compressed image file using dynamic range expansion (DRE) to form an image file x{circumflex over ( )}_e;

de-emphasizing said image file x{circumflex over ( )}_e using a filter D(ω

, G) to create a x{circumflex over ( )} image; and

using an Estimated Spectrum Adaptive Postfilter algorithm to post-filter said x{circumflex over ( )} image into an x{tilde over ( )} image.

6. The method of claim 5, wherein the ESAP algorithm comprises the steps of:

providing a decompressed image file;

providing dequantized DCT coefficients;

estimating 2-D pixel-adaptive bandwidths directly from said dequantized DCT coefficients; and

using said dequantized DCT coefficients to control a 2-D spatially-adaptive non-linear image post-filter.

7. The method of claim 6, further comprising the step of:

using an edge detector to control a 2-D spatially-adaptive non-linear image post-filter.

8. The method of claim 7, wherein the edge detector is Canny edge detector.

9. The method of claim 5, further comprising the step of:

using an Estimated Spectrum Adaptive Postfilter algorithm to post-filter said x{circumflex over ( )}_r image into an x{tilde over ( )}_r image.

10. The method of claim 5, further comprising the step of:

using an Estimated Spectrum Adaptive Postfilter algorithm to post-filter said x{circumflex over ( )}_e image into an x{tilde over ( )}_e image.