CONTENTS
The development of low-cost remote sensing systems is important to allow feasible use of images to gather information on small agriculture business, specially on developing coutries. However, images obtained through such systems have often particular features and high complexity elements that can hinder the analysis by image processing techniques used in satellite remote sensing.
In this project we investigated vegetation indices and segmentation techniques that allow detection of green coverage, gaps and degraded areas so that spatial data can be extracted from the images and used as information on precision farming and other applications
Segmentation of low-cost remote sensing images using vegetation indices and mean-shift IEEE GRSL, submitted Feb, 2012 (GRSL-00088-2012).
Matlab Code - Segmentation methods and evaluation. Requires EDISON library and wrapper, can be obtained on: Shai Bagon's Matlab Interface for EDISON
Last update Feb 7, 2012