Community Research and Development Information Service - CORDIS

Abstract

A range of classification approaches currently used in remote sensing application programs suffer from rd drawbacks related to the per pixel image treatment. Structural approaches to image analysis are commonly considered superior but will also not allow an adequate identification of agricultural crop types when only single date imagery is used. This stems from the fact that the most significant information for a correct identification of cover types in diverse agricultural systems can only be derived from the combination of images taken during the growing season. We describe the segmentation of a multi temporal data set which facilitates the combined use of structural image description and optimized radiometric information for a subsequent interpretation in terms of agricultural land cover categories.

Additional information

Authors: HILL J, JRC ISPRA ESTAB. (ITALY);MEGIER J JRC ISPRA ESTAB. (ITALY), JRC ISPRA ESTAB. (ITALY)
Bibliographic Reference: 1987 INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, ANN ARBOR, MICHIGAN (USA), MAY 18-21, 1987 WRITE TO CEC LUXEMBOURG, DG XIII/A2, POB 1907 MENTIONING PAPER E 33433 ORA
Availability: Can be ordered online
Record Number: 1989126021101 / Last updated on: 1989-01-01
Category: PUBLICATION
Available languages: en