Patent US2299768A successfully refreshed from Google Patents.
US4386366A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Adaptive type quantizer
Assignee: Fujifilm Holdings Corp
•
Filing: 1981-03-16
•
Granted: 1983-05-31
•
Expires: 2001-03-16
Abstract
Abstract A method for adaptively equalizing sampled input picture element values in which an estimation error obtained from a sampled input picture element value and the estimation value thereof are adaptively quantized by selecting one of a plurality of quantization characteristics. Quantization steps are determined in accordance with the desired quantization characteristics. Input groups of picture elements are divided into adjacent blocks and a quantization characteristic used for succeeding blocks is determined according to the distribution of quantization level values of preceding blocks. A quantization characteristic selecting unit selects a quantization characteristic set by the quantization characteristic determining unit from among a plurality of quantization characteristics. The estimation errors are quantized in a predetermined block using the quantization characteristic which is so selected.
Inventors
- 👤 Sumio Mori
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-24 09:18
CPC Classifications
H; H04; H04N; H04N19/00; H04N19/90; H04N19/10; H04N19/102; H04N19/124; H04N19/126; H04N19/134; H04N19/146; H04N19/149; H04N19/50; H04N19/60
Claims Analysis
Total Claims: 3
Claim 1
Independent Claim
287 words
A method for adaptively quantizing sampled input picture element values in which an estimation error obtained from a sampled input picture element value and the estimation value thereof are adaptively quantized by selecting one of a plurality of quantization characteristics, comprising the steps of: determining quantization steps for said quantization characteristics; dividing a group of input picture elements in to m adjacent blocks each having a plurality of picture elements for every line on an image, m quantization characteristic selection memories being provided in correspondence to said m blocks; setting with a quantization characteristic determining unit, according to the distribution of quantization level values provided when said estimation errors in one of said blocks are quantized with a selected quantization characteristic, a quantization characteristic to be used for the next block; and selecting with a quantization characteristic selecting unit a quantization characteristic set by said quantization characteristic determining unit out of said plurality of quantization characteristics, wherein for every block in a line to be quantized, said quantization characteristic to be used for the corresponding block in the next line is set by said quantization characteristic determining unit according to the distribution of said quantization level values in said block and the current quantization characteristic set to store a quantization characteristic symbol to be used at a predetermined address in said quantization characteristic selection memories, and wherein in quantizing each block, said quantization characteristic selecting unit refers to said quantization characteristic symbols stored in said predetermined address corresponding to said block in said quantization characteristic selection memories to select a quantization characteristic for said block; and quantizing said estimation errors in a predetermined block using said quantization characteristic which is selected by said quantization characteristic selecting unit.
Claim 2
Independent Claim
287 words
A method for adaptively quantizing sampled input picture element values in which an estimation error obtained from a sampled input picture element value and the estimation value thereof are adaptively quantized by selecting one of a plurality of quantization characteristics, comprising the steps of: determining quantization steps for said quantization characteristics; dividing a group of input picture elements in to m adjacent blocks each having a plurality of picture elements for every line on an image, m quantization characteristic selection memories being provided in correspondence to said m blocks; setting with a quantization characteristic determining unit, according to the distribution of quantization level values provided when said estimation errors in one of said blocks are quantized with a selected quantization characteristic, a quantization characteristic to be used for the next block; and selecting with a quantization characteristic selecting unit a quantization characteristic set by said quantization characteristic determining unit out of said plurality of quantization characteristics, wherein for every block in a line to be quantized, said quantization characteristic to be used for the corresponding block in the next line is set by said quantization characteristic determining unit according to the distribution of said quantization level values in said block and the current quantization characteristic set to store a quantization characteristic symbol to be used at a predetermined address in said quantization characteristic selection memories, and wherein in quantizing each block, said quantization characteristic selecting unit refers to said quantization characteristic symbols stored in said predetermined address corresponding to said block in said quantization characteristic selection memories to select a quantization characteristic for said block; and quantizing said estimation errors in a predetermined block using said quantization characteristic which is selected by said quantization characteristic selecting unit.
Claim 3
Dependent Claim
130 words
The method as claimed in claim 1 wherein said quantization characteristic determining unit, according to the distribution of quantization level values in one block, sets a quantization characteristic for the next block or for the same block in the next line, selects a quantization characteristic having a more coarse quantization step interval than a quantization characteristic used for the present block when a distribution d of quantization level values in a relevant block is larger than a threshold value α, a quantization characteristic having a finer quantization step interval than a quantization characteristic used for the present block when the distribution d is smaller than a threshold value β, and a quantization characteristic used for the present block when the distribution d is between said threshold values α and β.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1986-05-30 | 1986-11-29 | 1987-05-31 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1990-05-30 | 1990-11-29 | 1991-05-31 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1994-05-30 | 1994-11-29 | 1995-05-31 | $7700.00 | Window Closed / Lapsed |