The repository contains the implementation of different machine learning techniques on Hyperspectral and satellite Imagery analysis. Find more articles from here.
1. Basics - This notebook fatures:
- Introduction
- Downloading HSI
- Reading the hyperspecral image.
- Visualizing the bands of the hyperspectral image.
- Visualizing ground truth of the image.
- Extracting pixels of the hyperspectral image.
- Visualizing spectral signatures of the hyperspectral image.
2. Data Analysis - This notebook fatures data anlysis of the indian pines hyperspectral image:
- Visualizing pixels of the hyperspectral image.
- Bar plot w.r.t class labels of the hyperspectral image.
- Box Plot w.r.t the class labels and bands of hyperspecral image.
- Distribution Plot w.r.t the bands of hyperspecral image.
3.Exploratory Data Analysis (EDA) on Satellite Imagery Using EarthPy
4.Dimensionality Reduction
-
Check this article entitled Dimensionality Reduction in Hyperspectral Images using Python and code.
-
PCA + SVM - This notebook implements the following machine learning techniques on the indian pines dataset.
- Dimensionality Rreduction: The principal component analysis(PCA) is used to reduce the dimensions of the dataset.
- Classifier: The support vector machine(SVM) classifier is used to classsify the pixels of the HSI with classification report and the confusion matrix, classification map of the classifier is visualized.
-
Kernel PCA + SVM - This notebook implements the following machine learning techniques on the indian pines dataset.
- Dimensionality Rreduction: The Kernel principal component analysis(PCA) with 'rbf kernel' is used to reduce the dimensionality of the dataset.
- Classifier: The support vector machine(SVM) classifier is used to classsify the pixels of the HSI with classification report and the confusion matrix, classification map of the classifier is visualized.