Support Vector Machine and various methods of Multi-Spectral satellite image classification

摘要

Use of satellite images is one of the prominent methods for information about land coverage. Multi-Spectral satellite image is an appropriate source for providing this information. Classification of these Multi-Spectral images is an effective way to recover the information. This can be achieved based on the kinds of pattern models used, the types of information used, the manner in which they are applied to the image and the manner in which they partition the image into classes. Here, along with Support Vector Machine (SVM) algorithm, various other classification techniques are discussed and compared based on several parameters.