Integrating LANDSAT and SPOT images to improve landcover classification accuracy
The use of multi-sensor and multi-temporal data for land cover classification purposes is investigated. In particular a multilayer neural network trained by means of the Back Propagation algorithm is employed for classification experiments on remotely sensed images. The data set employed is composed of two co-registered images of the agricultural area surrounding the City of Valladolid (Spain) acquired on two different dates in June and July by two different satellites, SPOT and LANDSAT respectively, having different spectral and spatial resolutions; ground truth data were acquired during an in situ campaign carried out by the Spanish Ministry of Agriculture in 1993. Three different data sets (SPOT data, LANDSAT data and SPOT + LANDSAT data) are employed for the same land cover classification task in order to investigate the role of data integration and to compare the results.
Bibliographic Reference: Paper presented: 3rd International Workshop on Signal/Images processing, Manchester (GB), November 4-7, 1996
Availability: Available from (1) as Paper EN 40209 ORA
Record Number: 199710005 / Last updated on: 1997-02-03
Original language: en
Available languages: en