# US 5,347,311 A · Method and apparatus for unevenly encoding error images

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 5,347,311 A
- Document ID: US5347311A
- Status: Granted
- Kind: A
- Application number: 08/069,137
- Filed: 1993-05-28
- Granted: 1994-09-13
- Earliest claimed: 1993-05-28
- Assignees: Intel Corporation
- Inventors: Stuart J. Golin
- Web page: https://patentgrep.com/patent/US5347311A

## CPC

- H04N 19/126: Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
- H04N 19/12: Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
- H04N 19/61: in combination with predictive coding
- H04N 19/124: Quantisation
- H04N 19/20: using video object coding
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A method and apparatus for partially encoding an image. The image is divided into a plurality of blocks and one block is selected for processing. A boundary lying along the selected block that is associated with a discontinuity is also selected. The errors associated with the selected block are reduced unevenly by a reduction factor that increases the degree to which the errors are reduced as the distance between the errors and the boundary increases. The selected block is encoded in accordance with the reduced errors.

## Claims

### Claim 1

A method for partially encoding an error image, comprising the steps of:
(a) dividing said error image into a plurality of blocks;
(b) selecting one block from said plurality of blocks, said selected block having errors associated therewith;
(c) selecting a boundary lying along said selected block that is associated with a discontinuity;
(d) determining reduced errors by unevenly reducing said errors associated with said selected block, wherein said errors are reduced by a reduction factor that increasingly reduces said errors as a distance between said errors and said boundary increases; and
(e) encoding said selected block in accordance with said reduced errors.

### Claim 2 (depends on 1)

The method of claim 1, wherein said reduction factor varies linearly as said distance between said errors and said boundary increases.

### Claim 3 (depends on 1)

The method of claim 1, wherein said error image is a residual error image formed as a result of motion compensation.

### Claim 4 (depends on 1)

The method of claim 1, wherein said step (d) comprises the steps of:
(1) selecting a row from said selected block;
(2) determining said reduction factor in accordance with a distance between said selected row and said discontinuity;
(3) multiplying said selected row by said reduction factor; and
(4) repeating steps (1)-(3) for each row in said selected block.

### Claim 5

An apparatus for partially encoding an error image, comprising:
(a) means for dividing said error image into a plurality of blocks;
(b) a block selector for selecting one block from said plurality of blocks, said selected block having errors associated therewith;
(c) a block edge selector for selecting a boundary lying along said selected block that is associated with a discontinuity;
(d) means for determining reduced errors by unevenly reducing said errors associated with said selected block, wherein said errors are reduced by a reduction factor that increasingly reduces said errors as a distance between said errors and said boundary increases; and
(e) an encoder for encoding said selected block in accordance with said reduced errors.

### Claim 6 (depends on 5)

The apparatus of claim 5, wherein said reduction factor varies linearly as said distance between said errors and said boundary increases.

### Claim 7 (depends on 5)

The apparatus of claim 5, wherein said error image is a residual error image formed as a result of motion compensation.

### Claim 8 (depends on 5)

The apparatus of claim 5, wherein said means for determining reduced errors comprises:
(1) a row selector for selecting a row from said selected block;
(2) a reduction factor determinator for determining said reduction factor in accordance with a distance between said selected row and said discontinuity; and
(3) a multiplier for multiplying said selected row by said reduction factor.