High-spatial resolution multispectral and panchromatic satellite imagery for mapping perennial desert plants

Saad Alsharrah, David A. Bruce, Rachid Bouabid, Sekhar Somenahalli, Paul A. Corcoran

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

The use of remote sensing techniques to extract vegetation cover information for the assessment and monitoring of land degradation in arid environments has gained increased interest in recent years. However, such a task can be challenging, especially for medium-spatial resolution satellite sensors, due to soil background effects and the distribution and structure of perennial desert vegetation. In this study, we utilised Pleiades high-spatial resolution, multispectral (2m) and panchromatic (0.5m) imagery and focused on mapping small shrubs and low-lying trees using three classification techniques: 1) vegetation indices (VI) threshold analysis, 2) pre-built object-oriented image analysis (OBIA), and 3) a developed vegetation shadow model (VSM). We evaluated the success of each approach using a root of the sum of the squares (RSS) metric, which incorporated field data as control and three error metrics relating to commission, omission, and percent cover. Results showed that optimum VI performers returned good vegetation cover estimates at certain thresholds, but failed to accurately map the distribution of the desert plants. Using the pre-built IMAGINE Objective OBIA approach, we improved the vegetation distribution mapping accuracy, but this came at the cost of over classification, similar to results of lowering VI thresholds. We further introduced the VSM which takes into account shadow for further refining vegetation cover classification derived from VI. The results showed significant improvements in vegetation cover and distribution accuracy compared to the other techniques. We argue that the VSM approach using high-spatial resolution imagery provides a more accurate representation of desert landscape vegetation and should be considered in assessments of desertification.

Original languageEnglish
Title of host publicationEarth Resources and Environmental Remote Sensing/GIS Applications VI
EditorsUlrich Michel, Manfred Ehlers, Karsten Schulz, Konstantinos G. Nikolakopoulos, Daniel Civco
ISBN (Electronic)9781628418545
DOIs
Publication statusPublished - 1 Jan 2015
EventEarth Resources and Environmental Remote Sensing/GIS Applications VI - Toulouse, France
Duration: 22 Sep 201524 Sep 2015

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9644
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceEarth Resources and Environmental Remote Sensing/GIS Applications VI
CountryFrance
CityToulouse
Period22/09/1524/09/15

Keywords

  • classification
  • cover
  • Desertification
  • object-oriented
  • perennial
  • shadow
  • shrub
  • vegetation

Fingerprint Dive into the research topics of 'High-spatial resolution multispectral and panchromatic satellite imagery for mapping perennial desert plants'. Together they form a unique fingerprint.

Cite this