AVHRR Image Segmentation Using Modified Backpropagation Algorithm

Tao CHEN; Mikio TAKAGI

AVHRR Image Segmentation Using Modified Backpropagation Algorithm

Tao CHEN, Mikio TAKAGI

Full Text Views

0

Share
Cite this

Summary :

Analysis of satellite images requires classificatio of image objects. Since different categories may have almost the same brightness or feature in high dimensional remote sensing data, many object categories overlap with each other. How to segment the object categories accurately is still an open question. It is widely recognized that the assumptions required by many classification methods (maximum likelihood estimation, etc.) are suspect for textural features based on image pixel brightness. We propose an image feature based neural network approach for the segmentation of AVHRR images. The learning algoriothm is a modified backpropagation with gain and weight decay, since feedforward networks using the backpropagation algorithm have been generally successful and enjoy wide popularity. Destructive algorithms that adapt the neural architecture during the training have been developed. The classification accuracy of 100% is reached for a validation data set. Classification result is compared with that of Kohonen's LVQ and basic backpropagation algorithm based pixel-by-pixel method. Visual investigation of the result images shows that our method can not only distinguish the categories with similar signatures very well, but also is robustic to noise.

Publication: IEICE TRANSACTIONS on Information Vol.E77-D No.4 pp.490-497

Publication Date: 1994/04/25

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Issue on Neurocomputing)

Category: Image Processing

Cite this

Copy

Tao CHEN, Mikio TAKAGI, "AVHRR Image Segmentation Using Modified Backpropagation Algorithm" in IEICE TRANSACTIONS on Information, vol. E77-D, no. 4, pp. 490-497, April 1994, doi: .
Abstract: Analysis of satellite images requires classificatio of image objects. Since different categories may have almost the same brightness or feature in high dimensional remote sensing data, many object categories overlap with each other. How to segment the object categories accurately is still an open question. It is widely recognized that the assumptions required by many classification methods (maximum likelihood estimation, etc.) are suspect for textural features based on image pixel brightness. We propose an image feature based neural network approach for the segmentation of AVHRR images. The learning algoriothm is a modified backpropagation with gain and weight decay, since feedforward networks using the backpropagation algorithm have been generally successful and enjoy wide popularity. Destructive algorithms that adapt the neural architecture during the training have been developed. The classification accuracy of 100% is reached for a validation data set. Classification result is compared with that of Kohonen's LVQ and basic backpropagation algorithm based pixel-by-pixel method. Visual investigation of the result images shows that our method can not only distinguish the categories with similar signatures very well, but also is robustic to noise.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e77-d_4_490/_p

Copy

@ARTICLE{e77-d_4_490,
author={Tao CHEN, Mikio TAKAGI, },
journal={IEICE TRANSACTIONS on Information},
title={AVHRR Image Segmentation Using Modified Backpropagation Algorithm},
year={1994},
volume={E77-D},
number={4},
pages={490-497},
abstract={Analysis of satellite images requires classificatio of image objects. Since different categories may have almost the same brightness or feature in high dimensional remote sensing data, many object categories overlap with each other. How to segment the object categories accurately is still an open question. It is widely recognized that the assumptions required by many classification methods (maximum likelihood estimation, etc.) are suspect for textural features based on image pixel brightness. We propose an image feature based neural network approach for the segmentation of AVHRR images. The learning algoriothm is a modified backpropagation with gain and weight decay, since feedforward networks using the backpropagation algorithm have been generally successful and enjoy wide popularity. Destructive algorithms that adapt the neural architecture during the training have been developed. The classification accuracy of 100% is reached for a validation data set. Classification result is compared with that of Kohonen's LVQ and basic backpropagation algorithm based pixel-by-pixel method. Visual investigation of the result images shows that our method can not only distinguish the categories with similar signatures very well, but also is robustic to noise.},
keywords={},
doi={},
ISSN={},
month={April},}

Copy

TY - JOUR
TI - AVHRR Image Segmentation Using Modified Backpropagation Algorithm
T2 - IEICE TRANSACTIONS on Information
SP - 490
EP - 497
AU - Tao CHEN
AU - Mikio TAKAGI
PY - 1994
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E77-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 1994
AB - Analysis of satellite images requires classificatio of image objects. Since different categories may have almost the same brightness or feature in high dimensional remote sensing data, many object categories overlap with each other. How to segment the object categories accurately is still an open question. It is widely recognized that the assumptions required by many classification methods (maximum likelihood estimation, etc.) are suspect for textural features based on image pixel brightness. We propose an image feature based neural network approach for the segmentation of AVHRR images. The learning algoriothm is a modified backpropagation with gain and weight decay, since feedforward networks using the backpropagation algorithm have been generally successful and enjoy wide popularity. Destructive algorithms that adapt the neural architecture during the training have been developed. The classification accuracy of 100% is reached for a validation data set. Classification result is compared with that of Kohonen's LVQ and basic backpropagation algorithm based pixel-by-pixel method. Visual investigation of the result images shows that our method can not only distinguish the categories with similar signatures very well, but also is robustic to noise.
ER -