Investigating the utility of SPOT 5 imagery and artificial neural networks, in the identification and mapping of Acacia mearnsii within environments of varying complexity

Abstract

The impacts of invasive species on the environment, human health, and the economy continue to gain interest from public and private agencies, scientists, and the media. This study aimed to investigate the utility of SPOT 5 imagery and Artificial Neural Networks, in the identification and mapping of Acacia mearnsii within environments of varying complexity. Results showed that it is possible to identify and map Acacia mearnsii using SPOT 5 imagery, depending on the classification algorithm used. It was established that the neural network algorithms performed with greater success when compared to the performance of traditional classifiers, confirming other similar studies. The utility of the various classification algorithms was also investigated in terms of their applicability to environments of varying complexity. The neural networks once again, proved to be more successful and performed well in both the complex and relatively simple environments, indicating the robustness of the neural network algorithm.