Compression techniques for substantially lossless digital image data storage

• US 5,790,705 A
• Filed: 09/13/1996
• Issued: 08/04/1998
• Est. Priority Date: 09/13/1996
• Status: Expired due to Term

• Current Assignee: Apple Inc.
• Original Assignee: Apple Computer Incorporated (Apple Inc.)
• Inventors: Anderson, Eric C., Dalke, George W.
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US08/713,749
• Time in Patent Office: 690 Days
• Field of Search: 382/251, 382/244, 382/232, 341/95, 341/55, 341/50
• US Class Current: 382/244
• CPC Class Codes: G06T 9/00 Image coding bandwidth or redundancy reduction for static pictures H04N1/41; coding or decoding of static colour picture signals H04N1/64; methods or arrangements for coding, decoding, compressing or decompressing digital video signals H04N19/00
• CPC Class Codes: H03M 7/3053 Block-companding PCM systems

Abstract

A method for compressing digital image of more than a first predetermined number into the first predetermined number of bits in a substantially lossless manner includes determining a plurality of ranges in which input image data falls and comparing current image data with the plurality of ranges. The method further includes preserving a second predetermined number of bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits.

A system includes means for determining a plurality of ranges in which input image data falls and means for comparing current image data with the plurality of ranges, the means for comparing coupled to the means for determining. The system further includes means for preserving a second predetermined number of bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, the means for preserving coupled to the means for determining and the means for comparing.

Claims

1. A method for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the method comprising:

determining a plurality of ranges in which input image data falls;

comparing current image data with the plurality of ranges; and

preserving six bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, wherein when six bits of data are preserved, the step of determining ranges identifies a first range when the most significant bit (NSB) of image data is a one, a second range when the two MSBs of the image data are zero-one, a third range when the three MSBs of image data are zero-zero-one, and a fourth range when the three MSBs of the image data are a zero-zero-zero.

2. The method of claim 1 wherein the first predetermined number of bits comprises eight bits.

3. The method of claim 1 wherein the input image data comprises analog-to-digital convertor values from 0-1023.

4. The method of claim 1 wherein when the image data is in the first range, the method further comprises preserving the six bits of image data that immediately follow the MSB, and encoding the preserved six bits in the image value after a one-one sequence.

5. The method of claim 1 wherein when the image data is in the second range, the method further comprises preserving the six bits immediately following the two MSBs, and encoding the preserved six bits in the image value after a one-zero sequence.

6. The method of claim 1 wherein when the image data is in the third range, the method further comprises preserving the six bits of image data immediately following the three MSBs, and encoding the preserved six bits in the image value after a zero-one sequence.

7. The method of claim 1 wherein when the image data is in the fourth range, the method further comprises preserving the six bits of image data immediately following the three MSBs, and encoding the preserved six bits in the image value after a zero-zero sequence.

8. A method for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the method comprising:

determining a plurality of ranges in which input image data falls;

comparing current image data with the plurality of ranges; and

preserving seven bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, wherein when seven bits of data are preserved, the step of determining ranges identifies a first range when the three most significant bits (MSBs) of the image data are a zero-zero-zero, a second range when the two MSBs of the image data are zero-zero, a third range when the MSB of image data is zero, and a fourth range when the MSB of the image data is not a zero.

9. The method of claim 8 wherein when the image data is in the first range, the method further comprises preserving the seven bits of image data immediately following the three MSBs, and encoded the preserved seven bits as the seven least significant bits in the image value after a MSB code for a zero-zero sequence.

10. The method of claim 8 wherein when the image data is in the second range, the method further comprises preserving the seven bits of image data immediately following the two MSBs, and encoding the preserved seven bits as the seven least significant bits in the image value after a code for a zero-one sequence.

11. The method of claim 8 wherein when the image data is in the third range, the method further comprises preserving the seven bits of image data immediately following the MSB, and encoding the preserved seven bits as the seven least significant bits in the image value after a code for a one-zero sequence.

12. The method of claim 8 wherein when the image data is in the fourth range, the method further comprises preserving the seven bits of image data that are the seven MSBs, and encoding the preserved seven bits as the seven least significant bits in the image value after a code for a one-one sequence.

13. A method for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the method comprising:

determining a plurality of ranges in which input image data falls;

comparing current image data with the plurality of ranges; and

preserving eight bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, wherein when eight bits of image data are preserved, the method further comprises determining in-range and out-of-range conditions between previous image data and current image data.

14. The method of claim 13 wherein when an in-range condition occurs, a delta modulation encoding is performed.

15. The method of claim 13 wherein when an out-of-range condition occurs, the image value comprises a one followed by the bit values for the seven MSBs of the current image data.

16. A system for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the system comprising:

means for determining a plurality of ranges in which input image data falls;

means for comparing current image data with the plurality of ranges, the means for comparing coupled to the means for determining; and

means for preserving six bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, the means for preserving coupled to the means for determining and the means for comparing, wherein the means for determining and the means for comparing comprise a priority encoder coupled to a plurality of registers, and the means for preserving comprises at least one barrel shifter coupled to the priority encoder and the plurality of registers.

17. A system for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the system comprising:

means for determining a plurality of ranges in which input image data falls;

means for comparing current image data with the plurality of ranges, the means for comparing coupled to the means for determining; and

means for preserving seven bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, the means for preserving coupled to the means for determining and the means for comparing, wherein the means for determining and means for comparing comprise a 3-bit OR priority encoder coupled to a plurality of registers, and the means for preserving comprises a two-input select shifter coupled to the 3-bit priority OR encoder and the plurality of registers.

18. A system for compressing digital image of more than a first predetermined number of bits into the first predetermined number of bits in a substantially lossless manner, the system comprising:

means for determining a plurality of ranges in which input image data falls;

means for comparing current image data with the plurality of ranges, the means for comparing coupled to the means for determining; and

means for preserving eight bits of the current image data based upon which of the plurality of ranges the current image data falls to encode the current image data as an image value comprising the first predetermined number of bits, the means for preserving coupled to the means for determining and the means for comparing, wherein the means for determining and means for comparing comprises subtractor logic coupled to range checker logic, and the means for preserving comprises bit steering and assembly logic coupled to the subtractor logic and range checker logic.