No abstract

No abstract

The production of land cover maps using supervised classification algorithms is one of the most common applications of remote sensing. In this study, the effectiveness of supervised classification algorithms in land cover classification using ASTER data was evaluated in the Mankweng Area and its environs. The false colour composite image generated from combination of band 1, 2 and 3 in red, green and blue, respectively, was used to generate training classes for six land cover types (waterbody, forest, vegetation, Duiwelskloof leucogranite, Turfloop granite and built-up land). These were used to construct land cover maps using eight supervised classification algorithms: Maximum Likelihood, Minimum Distance, Support Vector Machine, Mahalanobis Distance, Parallelepiped, Neural Network, Spectral Angle Mapper and Spectral Information Divergence. To evaluate the effectiveness of the algorithms, the land cover maps were subjected to accuracy assessment to determine precision of the algorithms in accurately classifying the land cover types and level of confidence that can be attributed to the land cover maps. Most algorithms poorly performed in classifying spatially overlapping land cover types without abrupt boundaries. This indicates that the environmental conditions and distribution of land cover types can affect the performance of certain classification algorithms, and thus need to be considered prior to selection of algorithms. However, Support Vector Machine and Minimum Distance proved to be the two most effective algorithms as they provided better producer’s and user’s accuracy in the range of 80-100% for all land cover types, which represent good classification.

This study presents a simple approach of spatiotemporal change detection of vegetation cover based on analysis of time series remotely sensed images. The study was carried out at Thathe Vondo Area, which is characterised by episodic variation of vegetation gain and loss. This variation is attributable to timber and tea plantations and their production cycles, which periodically result in either vegetation gain or loss. The approach presented here was implemented on two ASTER images acquired in 2007 and 2017. It involved the combined use of band combination, unsupervised image classification and Normalised Difference Vegetation Index (NDVI) techniques. True colour composite (TCC) images for 2007 and 2017 were created from combination of bands 1, 2 and 3 in red, blue and green, respectively. The difference image of the TCC images was then generated to show the inconsistencies of vegetation cover between 2007 and 2017. For analytical simplicity and interpretability, the difference image was subjected to ISODATA unsupervised classification, which clustered pixels in the difference image into eight classes. Two ISODATA derived classes were interpreted as vegetation gain and one as vegetation loss. These classes were confirmed as regions of vegetation gain and loss by NDVI values of 2007 and 2017. In addition, the polygons of vegetation gain and loss regions were created and superimposed over the TCC images to further demonstrate the spatiotemporal vegetation change in the area. The vegetation change statistics show vegetation gain and loss of 10.62% and 2.03%, respectively, implying a vegetation gain of 8.59% over the selected decade.

No abstract