US5369503A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method of picture compression by auto-organisation of a neural network
Assignee: Thales SA
•
Filing: 1990-12-06
•
Granted: 1994-11-29
•
Expires: 2010-12-06
Abstract
Abstract The method of picture compression consists: in processing a current displaced pixel block in the picture to be compressed by using a network of N neurons, 1 to N, disposed in a single plane, each neuron having a number of inputs equal to the dimension of the blocks extracted from the picture, and such that, after a learning phase during which the network has auto-organised so that the neurons represent most probable input values, each neuron receives on its inputs, at the same instant, the pixels of the current block and outputs a signal, or potential, d j , which is positive and is all the lower the nearer is the state of the corresponding neuron j to the state of the inputs e i , in coding the current pixel block through a code associated with the neuron j whose potential d j is minimal. The invention applies, in particular, to the compression of television pictures.
Inventors
- 👤 Gilles Burel
- 👤 Isabelle Pottier
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-04 07:46
CPC Classifications
G; G06; G06T; G06T9/00; G06T9/002; G06F; G06F18/00; G06F18/20; G06F18/21; G06F18/213; G06F18/2137; G06N; G06N3/00; G06N3/02; G06N3/08; G06N3/088; G06T1/00; G06T1/20; H; H04; H04N; H04N19/00
Claims Analysis
Total Claims: 6
Claim 1
Independent Claim
223 words
Method of picture compression by autoorganisation of a neural network, characterised in that it comprises: a preprocessing during which pictures of identical type are partitioned into pixel blocks with dimensions smaller than those of the pictures, a phase of initialising a neural network comprising N neurons disposed in a single plane and having a number of inputs equal to the dimension of the blocks extracted from the pictures, during which values of coefficients, or weights, are randomly allocated to each connection linking a neuron j to an input ei, and an initial neighbourhood radius Vo determining the neighbour neurons of the neuron j is chosen a learning phase during which the neural network is fed successively with pixel blocks extracted randomly from the various pictures and the weights of the neural connections evolve by successive approximations following an algorithm simulating the behaviour of the visual cortex, through a computation of potentials which aims to elect one neuron with updating of its neighbours, until their values are as close as possible to most probable input values, a compression phase during which the source picture is partitioned into pixel blocks, each block being presented one after the other to the inputs of the neural network and coded as a function of the neuron of which the values of the weights are closest to the inputs.
Claim 2
Dependent Claim
223 words
Method of picture compression by autoorganisation of a neural network, characterised in that it comprises: a preprocessing during which pictures of identical type are partitioned into pixel blocks with dimensions smaller than those of the pictures, a phase of initialising a neural network comprising N neurons disposed in a single plane and having a number of inputs equal to the dimension of the blocks extracted from the pictures, during which values of coefficients, or weights, are randomly allocated to each connection linking a neuron j to an input ei, and an initial neighbourhood radius Vo determining the neighbour neurons of the neuron j is chosen a learning phase during which the neural network is fed successively with pixel blocks extracted randomly from the various pictures and the weights of the neural connections evolve by successive approximations following an algorithm simulating the behaviour of the visual cortex, through a computation of potentials which aims to elect one neuron with updating of its neighbours, until their values are as close as possible to most probable input values, a compression phase during which the source picture is partitioned into pixel blocks, each block being presented one after the other to the inputs of the neural network and coded as a function of the neuron of which the values of the weights are closest to the inputs.
Claim 3
Dependent Claim
62 words
Method according to claim 1, characterised in that with each block presented to the inputs of the neural network, the weights of the neuron whose state is closest to the state of the inputs and the weights of all the neighbour neurons are modified so that their values approximate the values of the inputs, the weights of the other neurons being unchanged.
Claim 4
Dependent Claim
27 words
Method according to claim 2, characterised in that the neighbourhood radius between a neuron j and the other neurons of the network diminishes during the learning phase.
Claim 5
Dependent Claim
14 words
Method according to claim 3, characterised in that the algorithm is the Kohonen algorithm.
Claim 6
Dependent Claim
32 words
Method according to any one of the preceding claims, characterised in that the neural network is looped in a ring, so that all the neurons have the same number of neighbour neurons.
References Cited (Backward Citations)
US5041916A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1997-11-28 | 1998-05-30 | 1998-11-29 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2001-11-28 | 2002-05-30 | 2002-11-29 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2005-11-28 | 2006-05-30 | 2006-11-29 | $7700.00 | Window Closed / Lapsed |