Integrated land cover mapping from satellite imagery using artificial neural networks
The automatic mapping of land cover from satellite imagery requires optimal classification and spatial generalisation procedures. Here the use of functional link neural networks is described. These are based on a flat perceptron net with an augmented feature vector, to generate high accuracy classification products. They can then be trained more rapidly than multi-layer perceptrons. The network output is used to fix land cover class area statistics which control a low-level generalisation procedure based on a combined iterative majority filtering and reduced class growing procedure.
Bibliographic Reference: Paper presented: SPIE Aerospace and Remote Sensing Symposium, Orlando (US), April 12-16, 1993
Availability: Available from (1) as Paper EN 37407 ORA
Record Number: 199310578 / Last updated on: 1994-11-29
Original language: en
Available languages: en