Author

Emil Bóasson

Date of Award

9-1981

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Geography

Supervisor

Dr. P.J. Howarth

Abstract

Several remote sensing systems have been evaluated to determine their capabilities for providing land cover/use information. The systems studied were LANDSAT Multispectral Scanner (MSS) visual and digital data, LANDSAT Return Beam Vidicon (RBV) imagery and Synthetic Aperture Radar (SAR) imagery, Standard panchromatic aerial photography provided the ground information against which the imagery was evaluates. The study was carried out in the Hamilton-Wentworth Region of southern Ontario.

The visual data were analyzed using transparencies in a Zoom Transfer Scope to produce a land cover/use map at a scale of 1:50,000. The digital data were preprocessed using the Digital Image Correction System (DICS) at the Canada Centre for Remote Sensing. Land cover/use maps were produced from the data using both a maximum likelihood classifier and a parallelepiped classifier.

It was found that the visual data gave a higher classification accuracy than the digital data. With an increasing number of classes, the accuracy of classification decreases.

Using LANDSAT data from two dates, several digital enhacements were tested for their change detection capability. The overlay of Band 5 data from both dates, combined to give a colour image, was the most effective way to identify areas of change.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Geography Commons

Share

COinS