Cover ImageCover Image
This Landsat Thematic Mapper (TM) images shows portions of the Florida Everglades National Park and Florida Keys. The background image is a false color infrared composite TM Image (Bands 4, 3, 2 as RGB). The various shades of red show vegetation types and vigor. The left inset shows an RGB with band-ratios 4/3, 3/1 and 4/5. Color ratio composites such as this are used to enhance selected spectral characteristics. The right inset is a Minimum Noise Fraction (MNF) transformation; the MNF transform is used to model the noise in the data to allow quantitative measures of spectral separability. These data are being used to study various pigments associated with concentrations of green and blue green algae, and phytoplankton in Florida Bay. The TM data was imported and processed using ENVI (the Environment for Visualizing Images) from Research Systems. For more information, contact Research Systems, Inc., 303-786-9900; envi@rsinc.com; www.rsinc.com.
Peer Reviewed Articles
485-491 Rule-Based Classification of Water
in Landsat MSS Images Using the
Variance Filter
Paul A. Wilson
The rules-based algorithm does not generate any false positives, whereas the threshold algorithm misclassifies many shadow pixels as water.
493-500 Detecting Subpixel Woody Vegetation in Digital Imagery Using Two Artificial
Intelligence Approaches
Patricia G. Foschi and Deborah K. Smith
A rule-based scheme, based on a machine-vision approach, was developed and a back-propagation neural network was employed to classify subpixel woody vegetation in simulated SPOT HRV imagery.
501-514 Performance of a Neural Network: Mapping Forests Using GIS and Remotely Sensed
Data
A.K. Skidmore, B.J. Turner, W. Brinkhof, and E. Knowles
The neural-network approach does not offer significant advantages over conventional classification schemes for mapping eucalypt forestss from Landsat TM and ancillary GIS data at the Anderson Level III forest type level.
515-521 Texture Analysis of Tropical Rain Forest Infrared Satellite Images
Robert Riou and Frederique Seyler
A Fourier transform algorithm for processing an entire image aimed at regional studies of texture networks is proposed, and applications to structural geology, forest-type descrimination, and soil studies in the tropics are briefly evaluated.
523-533 Multisource Classification of Complex
Rural Areas by Statistical and Neural-Network Approaches
L. Bruzzone, C. Conese, F. Maselli, and F. Roli
A statistical and neural-network classification approach are applied to a multisource data set related to Italian complex rural areas, and their classification performances are evaluated and compared in a quantitative and detailed way.
535-544 The Effect of Neural-Network Structure
on a Multispectral Land-Use/Land-Cover Classification
Justin D. Paola and Robert A. Schowengerdt
The size of the hidden layer in a neral network must be determined by trial and error, and the random initial weight settings result in different paths for the training procedure, making the netework a non-deterministic classifier.