Using Remote Sensing to Identify Illegal Crop Plantation
Identification of Cannabis plantation
By Manuel Lukas
The remoteness of some outdoor cannabis plantations tends to decrease the chance of detection by law enforcement agencies. In NSW, the NSW Police Force have been using the National Parks and Wildlife helicopter to detect cannabis plantation (ABCI, 1999).
The research aims to analyse a historical site of cannabis plantation and to observe the effectiveness of different satellite scanner imagery on detecting the cannabis plantation and hence to anticipate the ability to determined a specific spectral signature of cannabis through the image analysis of different satellite imagery.
Satellite image from Landsat 5 TM (Thematic Mapper) will be used in the image analysis to classify and identify cannabis plantation. A numbers of cannabis spectral signatures have been observed using a digital spectrometer, the best spectra in the identification of plantation will be concluded as the specific spectral signature of cannabis.
Several method of image analysis such as; supervised classification, Normalised Difference Vegetation Index, change detection, and Spectral Angle Mapper classifications will be applied. The results of the image analysis will be analysed to see how well the satellite imagery can detect the cannabis plantation from the other vegetation and the geographical feature around the site.
The type of constraints in identifying cannabis plantation using remote sensing are; plantation grown under high trees’ canopy and the availability of the image in good condition (i.e. without clouds or any other events such as bushfire occurring).
For more information, please contact Project Supervisor
Dr. Linlin Ge
School of Surveying and Spatial Information Systems
University of New South Wales
UNSW SYDNEY NSW 2052