US6424737B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus of compressing images using localized radon transforms
Assignee: Sony Corp
•
Filing: 2001-01-22
•
Granted: 2002-07-23
•
Expires: 2021-01-22
Abstract
Abstract A method and an apparatus of compressing data. The method and apparatus include constructing a neural network having a specific geometry using a finite and discrete Radon transform. The data is then fed through the neural network to produce a transformed data stream. The transformed data stream is thresholded. A fixed input signal is fed back through the neural network to generate a decoding calculation of an average value. The thresholded data stream is entropy encoded.
Inventors
- 👤 Hawley K. Rising, III
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-30 15:33
CPC Classifications
G; G06; G06T; G06T9/00; G06T9/002; G06N; G06N3/00; G06N3/02; G06N3/04; G06N3/0495; G06N3/0499; G06N3/08; G06N3/082; G06N3/084; H; H03; H03M; H03M7/00; H03M7/30; H04; H04N; H04N19/00; H04N19/60; H04N19/63
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
63 words
A method of compressing data, the method comprising: constructing a neural network having a specific geometry using a finite and discrete Radon transform; feeding the data through the neural network to produce a transformed data stream; thresholding the transformed data stream; feeding back a fixed input signal to generate a decoding calculation of an average value; and entropy encoding the thresholded data stream.
Claim 2
Independent Claim
63 words
A method of compressing data, the method comprising: constructing a neural network having a specific geometry using a finite and discrete Radon transform; feeding the data through the neural network to produce a transformed data stream; thresholding the transformed data stream; feeding back a fixed input signal to generate a decoding calculation of an average value; and entropy encoding the thresholded data stream.
Claim 3
Dependent Claim
32 words
The method of claim 1 further comprising dividing the thresholded data stream to produce more than one stream of thresholded data if the data is to be stored in a distributed mode.
Claim 4
Dependent Claim
19 words
The method of claim 1 wherein designing a neural network comprises designing a neural network based on predetermined criteria.
Claim 5
Dependent Claim
48 words
The method of claim 3 wherein the predetermined criteria is selected from the group consisting of simplicity of encoding, simplicity of decoding, natural geometries suggested by subject matter of the data, and natural geometries suggested by network architecture of the network in which the data will be stored.
Claim 6
Dependent Claim
15 words
The method of claim 1 further comprising: storing the entropy encoded data in distributed mode.
Claim 7
Dependent Claim
17 words
The method of claim 1 wherein thresholding the transformed data stream comprises thresholding based on predetermined criteria.
Claim 8
Dependent Claim
27 words
The method of claim 6 wherein the predetermined criteria is selected from the group consisting of quality of features to be preserved or a desired compression ratio.
Claim 9
Dependent Claim
23 words
The method of claim 1 wherein thresholding the transformed data stream includes removing components of the data stream above a predetermined maximum frequency.
Claim 10
Independent Claim
73 words
A system for compressing data, the system comprising: means for constructing a neural network having a specific geometry using a finite and discrete Radon transform; means for feeding the data through the neural network to produce a transformed data stream; means for thresholding the transformed data stream; means for feeding back a fixed input signal to generate a decoding calculation of an average value; and means for entropy encoding the thresholded data stream.
Claim 11
Independent Claim
73 words
A system for compressing data, the system comprising: means for constructing a neural network having a specific geometry using a finite and discrete Radon transform; means for feeding the data through the neural network to produce a transformed data stream; means for thresholding the transformed data stream; means for feeding back a fixed input signal to generate a decoding calculation of an average value; and means for entropy encoding the thresholded data stream.
Claim 12
Dependent Claim
76 words
A computer readable medium comprising instructions, which when executed on a processor, perform a method for compressing data, the method comprising: constructing a neural network having a specific geometry using a finite and discrete Radon transform; feeding the data through the neural network to produce a transformed data stream; thresholding the transformed data stream; feeding back a fixed input signal to generate a decoding calculation of an average value; and entropy encoding the thresholded data stream.
Claim 13
Dependent Claim
76 words
A computer readable medium comprising instructions, which when executed on a processor, perform a method for compressing data, the method comprising: constructing a neural network having a specific geometry using a finite and discrete Radon transform; feeding the data through the neural network to produce a transformed data stream; thresholding the transformed data stream; feeding back a fixed input signal to generate a decoding calculation of an average value; and entropy encoding the thresholded data stream.
Claim 14
Independent Claim
117 words
An apparatus for compressing data, the apparatus comprising: a neural network constructor that uses a finite and discrete Radon transform to construct a neural network having a specific geometry; a module to feed the data through the neural network to produce a transformed data stream, the module coupled to the neural network constructor; a thresholder to threshold the data stream, the thresholder coupled to the module; a fixed input signal feeder to feed back a fixed input signal to generate a decoding calculation of an average value, the fixed input signal feeder coupled to the thresholder; and an entropy encoder to entropy encode the thresholded data stream, the entropy encoder coupled to the fixed input signal feeder.
Claim 15
Independent Claim
117 words
An apparatus for compressing data, the apparatus comprising: a neural network constructor that uses a finite and discrete Radon transform to construct a neural network having a specific geometry; a module to feed the data through the neural network to produce a transformed data stream, the module coupled to the neural network constructor; a thresholder to threshold the data stream, the thresholder coupled to the module; a fixed input signal feeder to feed back a fixed input signal to generate a decoding calculation of an average value, the fixed input signal feeder coupled to the thresholder; and an entropy encoder to entropy encode the thresholded data stream, the entropy encoder coupled to the fixed input signal feeder.
Claim 16
Dependent Claim
40 words
The apparatus of claim 11 further comprising a divider to divide the thresholded data stream such that the thresholded data stream produces more than one stream of thresholded data if the data is to be stored in a distributed mode.
Claim 17
Dependent Claim
21 words
The apparatus of claim 11 wherein the neural network constructor is configured to design a neural network based on predetermined criteria.
Claim 18
Dependent Claim
48 words
The apparatus of claim 13 wherein the predetermined criteria is selected from the group consisting of simplicity of encoding, simplicity of decoding, natural geometries suggested by subject matter of the data, and natural geometries suggested by network architecture of the network in which the data will be stored.
Claim 19
Dependent Claim
19 words
The apparatus of claim 11 further comprising a storage module to store the entropy encoded data in distributed mode.
Claim 20
Dependent Claim
20 words
The apparatus of claim 11 wherein the thresholder is configured to threshold the transformed data stream based on predetermined criteria.
Claim 21
Dependent Claim
27 words
The apparatus of claim 16 wherein the predetermined criteria is selected from the group consisting of quality of features to be preserved and a desired compression ratio.
Claim 22
Dependent Claim
22 words
The apparatus of claim 11 wherein the thresholder is configured to remove components of the data stream above a predetermined maximum frequency.
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) | 2005-07-22 | 2006-01-21 | 2006-07-23 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2009-07-22 | 2010-01-21 | 2010-07-23 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2013-07-22 | 2014-01-21 | 2014-07-23 | $7700.00 | Window Closed / Lapsed |