Remote sensing images segmentation by Deriche's filter and neural network

Raphael K. Koffi; Basel Solaiman; Marie-Catherine Mouchot

doi:10.1117/12.196761

30 December 1994 Remote sensing images segmentation by Deriche's filter and neural network

Raphael K. Koffi, Basel Solaiman, Marie-Catherine Mouchot

Proceedings Volume 2315, Image and Signal Processing for Remote Sensing; (1994) https://doi.org/10.1117/12.196761
Event: Satellite Remote Sensing, 1994, Rome, Italy

Abstract

An image segmentation method for remote sensing data using hybride techniques is proposed. Edge detection approach for segmentation is considered in our study. Our aim is to integrate segmentation results in further processing namely classification. Images of the land from satellite are often corrupted by noise. On one hand, optimal edge detectors insure good noise immunity. On the other hand, the multi-layer perceptron (MLP) neural network has been found to be suited for classification. So we propose to combine these two techniques to improve segmentation process. Satellites for remote sensing provide several images for the same area, coded differently according to spectral bands. In order to bear in mind spectral and spatial information, neighborhood relation of pixels and different bands are taken into consideration during the classification realized by the neural network. Samples which constituate the training set for the MLP are selected from the third, fourth and fifth band and represent edge and non-edge patterns. Each sample vector is composed of the value of a current pixel in the local maxima image (enhancement image obtained by Deriche's filter) and its 8 nearest neighbors. The proposed method provides satisfactory results for our application and compared to other similar methods.

Citation Download Citation

Raphael K. Koffi, Basel Solaiman, and Marie-Catherine Mouchot "Remote sensing images segmentation by Deriche's filter and neural network", Proc. SPIE 2315, Image and Signal Processing for Remote Sensing, (30 December 1994); https://doi.org/10.1117/12.196761

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available