Hyperspectral Image Classification using Minimum Noise Fraction and Random Forest

Remote sensing technology is improving day by day which has also increased the uses of hyperspectral imaging tremendously. Exact classification of ground features from hyperspectral images is an important and a popular research area and also have attracted a widespread of attention. A good classification results are achieved by many methods for the classification of hyperspectral imaging. This paper focuses on classification of hyperspectral image using different machine learning techniques like support vector machine (SVM), random forest (RF), polynomial logistic regression (LR), K-Nearest Neighbour (KNN) and decision tree (DT). Principal Component Analysis (PCA) and Minimum Noise Fraction (MNF) have been used to reduce the unnecessary and noisy bands present in the dataset.