Abstract:
The invasive plant known as bugweed (Solanum mauritianum) is a notorious invader of
forestry plantations in the eastern parts of South Africa. Not only is bugweed considered to be
one of five most widespread invasive alien plant (IAP) species in the summer rainfall regions
of South Africa but it is also one of the worst invasive alien plants in Africa. It forms dense
infestations that not only impacts upon commercial forestry activities but also causes
significant ecological and environment damage within natural ecotones. Effective weed
management efforts therefore require new and robust approaches to accurately detect; map
and monitor weed distribution in order to mitigate the impact on forestry operations. In this
regard, support vector machines (SVM) offer a promising alternative to conventional machine
learning and pattern recognition approaches to weed detection and mapping using remote
sensing. The main objective of this research was to determine the utility of using a recursive
feature elimination support vector machine (SVM-RFE) based approach with a 272-waveband
AISA Eagle image to detect and map the presence of co-occuring bugweed within mature
Pinus patula compartments in KwaZulu Natal. The SVM-RFE approach required only 17
optimal bands from the original 272 band image to produce a classification accuracy of 93%
and True Skills Statistic of 0.83. Results from this study indicate that (1) there is definite
potential for using SVMs for accurate detection and mapping of bugweed species in
commercial plantations and (2) it is not necessary to use the entire 272-band dataset to
accurately detect bugweed occurrence as the SVM-RFE approach will identify an optimal
subset of wavebands for weed detection enabling substantially improved data processing and
analysis.
