Off-campus South Dakota State University users: To download campus access theses, please use the following link to log into our proxy server with your South Dakota State University ID and password.

Non-South Dakota State University users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Document Type

Thesis - University Access Only

Award Date

2014

Degree Name

Master of Science (MS)

Department / School

Plant Science

First Advisor

Dennis Helder

Abstract

Developing a BRDF model for satellite images is a well-known and ongoing topic in remote sensing engineering. A BRDF model is needed to consider for getting the true reflectance from remotely sensed objects. There are many BRDF models, which are developed and modified based on the target and user specifications. The aim of this thesis is to use an appropriate BRDF model for sensing the vegetated area’s spectra. To fulfill the aim, the questions arise of how many BRDF models exist, which model suits what type of target and that model’s pros and cons. The response to those questions is studied here. After that, this thesis discusses the reason to pick Roujean and PROSAIL model specifically for implementation and their validation. The validation result and the literature reviews say that the PROSAIL model is more appropriate for vegetation than the Roujean model regarding the hotspot effect. The thesis also analyses the way in which the PROSAIL model can be used in more perfect way for remote sensing satellite images, in the case of vegetated sites. This study was done in detail for Soybeans. Pilot investigations of Wheat and Corn are also experimented with but not included in detail. To find an exact methodology for implementing the PROSAIL model, this study tries to match the simulated spectra from the model with the observed spectra from outdoor field measurements. It concludes, for getting plant true reflectance, it is necessary to be exact with the plant bio-physical parameters which changes over growth. This concept is a very important facet of this thesis. To validate this idea, an indoor plantation was initiated and plant spectral measurements were done in the SDSU Optics lab. From the comparison of simulated and observed indoor spectra and also from the careful observation of plant growth an attempt was made to link model parameters to an actual plant set. However, results were limited due to the indoor plantation under artificial lighting conditions not being dense enough for reliable absolute measurements. This being said, it should be noted that results from the plants and from modeling were sufficient to undertake a sensitivity analysis to determine which parameter is more responsible to change the reflectance value in specific band. So, from the analysis and work, it is found that this model works well though it is too complex and for implementing this model with full accuracy, the plant bio-physical parameters should be based on outdoor field measurements for successful implementation of the PROSAIL model. If this is done, then the model can be reliably used to remove the BRDF effect from satellite images.

Library of Congress Subject Headings

Artificial satellites in agriculture
Agriculture -- Remote sensing
Vegetation mapping -- Remote sensing

Description

Includes bibliographical references (pages 127-129)

Format

application/pdf

Number of Pages

151

Publisher

South Dakota State University

Rights

In Copyright - Non-Commercial Use Permitted
http://rightsstatements.org/vocab/InC-NC/1.0/

Download

Share

COinS