Super-Resolution Land Cover Pattern Prediction Using a Hopfield Neural Network
Super-Resolution Land Cover Pattern Prediction Using a Hopfield Neural Network
Landscape pattern represents a key variable in management and understanding of the environment, as well as driving many environmental models. Remote sensing can be used to provide information on the spatial pattern of land cover features, but analysis and classification of such imagery suffers from the problem of class mixing within pixels. Fuzzy classification techniques can estimate the class composition of image pixels. However, their output provides no indication of how such classes are distributed spatially within the instantaneous field of view represented by the pixel. Techniques to provide an improved spatial representation of land cover targets larger than the size of a pixel have been developed, however, the mapping of sub-pixel scale land cover features has yet to be investigated. We recently described the application of a Hopfield neural network technique to super-resolution mapping of land cover features larger than a pixel (Tatem et al., 2000), using information of pixel composition determined from fuzzy classification, and (was but) now show how our approach can be extended in a new way (added) to predict the spatial pattern of sub-pixel scale features. The network converges to a minimum of an energy function defined as a goal and several constraints. Prior information on the typical spatial arrangement of the particular land cover types is incorporated into the energy function as a constraint. This produces a prediction of the spatial pattern of the land cover in question, at the sub-pixel scale. The technique is applied to synthetic and simulated Landsat TM imagery, and compared to results of an existing super-resolution target identification technique. Results show that the new approach (was Hopfield neural network) represents a simple, robust and efficient tool for super-resolution land cover pattern prediction from remotely sensed imagery.
1-14
Tatem, A.J.
ab877a1f-6cc7-4eae-8c15-bb299417223f
Lewis, H.G.
e9048cd8-c188-49cb-8e2a-45f6b316336a
Atkinson, P.M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Nixon, M.S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
January 2002
Tatem, A.J.
ab877a1f-6cc7-4eae-8c15-bb299417223f
Lewis, H.G.
e9048cd8-c188-49cb-8e2a-45f6b316336a
Atkinson, P.M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Nixon, M.S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Tatem, A.J., Lewis, H.G., Atkinson, P.M. and Nixon, M.S.
(2002)
Super-Resolution Land Cover Pattern Prediction Using a Hopfield Neural Network.
Remote Sensing of Environment, 79 (1), .
(doi:10.1016/S0034-4257(01)00229-2).
Abstract
Landscape pattern represents a key variable in management and understanding of the environment, as well as driving many environmental models. Remote sensing can be used to provide information on the spatial pattern of land cover features, but analysis and classification of such imagery suffers from the problem of class mixing within pixels. Fuzzy classification techniques can estimate the class composition of image pixels. However, their output provides no indication of how such classes are distributed spatially within the instantaneous field of view represented by the pixel. Techniques to provide an improved spatial representation of land cover targets larger than the size of a pixel have been developed, however, the mapping of sub-pixel scale land cover features has yet to be investigated. We recently described the application of a Hopfield neural network technique to super-resolution mapping of land cover features larger than a pixel (Tatem et al., 2000), using information of pixel composition determined from fuzzy classification, and (was but) now show how our approach can be extended in a new way (added) to predict the spatial pattern of sub-pixel scale features. The network converges to a minimum of an energy function defined as a goal and several constraints. Prior information on the typical spatial arrangement of the particular land cover types is incorporated into the energy function as a constraint. This produces a prediction of the spatial pattern of the land cover in question, at the sub-pixel scale. The technique is applied to synthetic and simulated Landsat TM imagery, and compared to results of an existing super-resolution target identification technique. Results show that the new approach (was Hopfield neural network) represents a simple, robust and efficient tool for super-resolution land cover pattern prediction from remotely sensed imagery.
Text
tatem_rse.pdf
- Other
More information
Submitted date: 20 November 2000
Published date: January 2002
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 256187
URI: http://eprints.soton.ac.uk/id/eprint/256187
ISSN: 0034-4257
PURE UUID: b71f8b30-46f8-4766-b5c2-6cbdf29bd71f
Catalogue record
Date deposited: 05 Mar 2004
Last modified: 15 Mar 2024 02:54
Export record
Altmetrics
Contributors
Author:
A.J. Tatem
Author:
P.M. Atkinson
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics