Method to derive edge extensions for wavelet transforms and inverse wavelet transforms

• US 5,828,849 A
• Filed: 05/14/1996
• Issued: 10/27/1998
• Est. Priority Date: 05/14/1996
• Status: Expired due to Term

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: LSI Logic Corporation (Broadcom, Inc.)
• Inventors: Vafai, Manoucher, Lempel, Mody, Ramachandran, Loganath
• Primary Examiner(s): Voeltz, Emanuel Todd
• Assistant Examiner(s): ASSOUAD, PATRICK J
• Application Number:
  US08/645,574
• Time in Patent Office: 896 Days
• Field of Search: 395/200.18, 395/200.77, 382/232, 382/248, 382/266, 382/276, 348/395, 348/387, 348/463
• US Class Current: 709/247
• CPC Class Codes: H04N 19/63 using sub-band based transform, e.g. wavelets
• CPC Class Codes: H04N 19/649 the transform being applied to non rectangular image segments

Abstract

A method derives edge extensions for wavelet transforms and inverse wavelet transforms of two-dimensional images. The method overcomes the necessity of side computations by treating the two-dimensional matrix of values as a one-dimensional array of values. The use of a one-dimensional array reduces the required flushing and loading of registers by allowing the flushing and loading to be performed in between frames, rather than in between rows or columns of the matrix.

Claims

1. In a wavelet compression-decompression system for multimedia computing, a method for deriving edge extensions for a two-dimensional matrix of values in multiple rows comprising:

forming a one-dimensional array from the two-dimensional matrix by sequencing ends of the rows together;

deriving the edge extensions for each value from adjacent values in the one-dimensional array.

2. The method of claim 1, wherein the ends of the rows are sequenced in raster-scan order to form the one-dimensional array.

3. The method of claim 1, wherein the end of the rows are sequenced in zig-zag order to form the one-dimensional array.

4. The method of claim 1, wherein the one-dimensional array is cyclical.

5. The method of claim 2, wherein the one-dimensional array is cyclical.

6. The method of claim 5, wherein the edge extensions are used to determine future and back values for a wavelet transform.

7. The method of claim 5, wherein the edge extensions are used to determine future and back values for an inverse wavelet transform.

9. The method of claim 1, wherein the preceding row is the last row if the present row is the first row, and wherein the next row is the first row if the present row is the last row.

10. The method of claim 1, wherein the first end is a left end and the second end is a right end.

8. In a wavelet compression-decompression system for multimedia computing, a method for deriving edge extensions for a two-dimensional matrix of values in multiple rows comprising:

determining whether processing a present value requires a back value because the present value is sufficiently close to a first end of a present row;

determining whether processing the present value requires a future value because the present value is sufficiently close to a second end of the present row;

obtaining the back value from the second end of a preceding row, if processing the present value requires the back value; and

obtaining the future value from the first end of a next row, if processing the present value requires the back value.

11. In a wavelet compression or decompression system for multimedia computing, a method for deriving edge extensions for a two-dimensional matrix of values in multiple rows comprising:

determining whether processing a present value requires a back value because the present value is sufficiently close to a first end of a present row;

determining whether processing the present value requires a future value because the present value is sufficiently close to a second end of the present row;

obtaining the back value from the first end of a preceding row, if processing the present value requires the back value; and

obtaining the future value from the second end of a next row, if processing the present value requires the back value.

12. The method of claim 11, wherein the preceding row is the last row if the present row is the first row, and wherein the next row is the first row if the present row is the last row.