Pca and random forest improve hyperspectral image classification accuracy
Dimensionality Reduction for Hyperspectral Image Classification
Computer Vision and Pattern Recognition
Summary
Hyperspectral images contain lots of data that can be hard to manage and analyze. The authors look at ways to reduce this data without losing important details, using techniques like PCA and LDA. They also compare different computer methods to classify the images. Their tests show that using PCA together with a Random Forest classifier gives the best results for identifying features in satellite images.
What this means in practice
- •For satellite data analysts: Improve land cover classification accuracy in satellite imagery by combining PCA with Random Forest classification methods.
- •For environmental monitoring teams: Use efficient dimensionality reduction and classification to speed up identifying changes in vegetation and natural resources from hyperspectral data.
Authors
Mohamed Cherifi, Ammar Mesloub, Mohammed Nabil El Korso, Tayeb Touhami, Abdennour Hacine Gharbi
Abstract
This paper addresses the issue of supervised classification in the context of hyperspectral satellite images. It deals with two fundamental aspects: dimensionality reduction of data and the selection of appropriate supervised classification techniques. Firstly, we delve into dimensionality reduction, a critical step in simplifying the management of hyperspectral data. The reduction aims to decrease complexity in terms of memory and computing time. We examine two commonly used methods: Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). Subsequently, we explore the selection of the most suitable supervised classification algorithms for hyperspectral images. We compare the performance of three methods: K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forest (RF) using real hyperspectral data. The results highlight that the combination of PCA and RF yields the highest overall accuracy and Kappa coefficient.