# Hyperspectral Image Classification with Markov Random Fields and a Convolutional Neural Network

## TL;DR

This paper introduces a novel hyperspectral image classification method that combines CNNs and Markov Random Fields within a Bayesian framework, effectively integrating spectral and spatial data for improved accuracy.

## Contribution

The paper proposes a new supervised classification algorithm that unifies CNN-based posterior estimation with spatial smoothness priors in a Bayesian setting, enhancing hyperspectral image analysis.

## Key findings

- Outperforms state-of-the-art methods on synthetic and benchmark datasets.
- Effectively integrates spectral and spatial information for better classification.
- Achieves higher accuracy in hyperspectral image classification tasks.

## Abstract

This paper presents a new supervised classification algorithm for remotely sensed hyperspectral image (HSI) which integrates spectral and spatial information in a unified Bayesian framework. First, we formulate the HSI classification problem from a Bayesian perspective. Then, we adopt a convolutional neural network (CNN) to learn the posterior class distributions using a patch-wise training strategy to better use the spatial information. Next, spatial information is further considered by placing a spatial smoothness prior on the labels. Finally, we iteratively update the CNN parameters using stochastic gradient decent (SGD) and update the class labels of all pixel vectors using an alpha-expansion min-cut-based algorithm. Compared with other state-of-the-art methods, the proposed classification method achieves better performance on one synthetic dataset and two benchmark HSI datasets in a number of experimental settings.