Land Cover Classification from Remote Sensing Images Based on Multi-Scale Fully Convolutional Network

TL;DR

This paper introduces a Multi-Scale Fully Convolutional Network that leverages multi-scale kernels to improve land cover classification accuracy from satellite imagery.

Contribution

It proposes a novel MSFCN architecture that effectively captures multi-scale features for better land cover classification.

Findings

Enhanced classification accuracy on satellite images

Effective multi-scale feature extraction demonstrated

Improved performance over existing methods

Abstract

In this paper, a Multi-Scale Fully Convolutional Network (MSFCN) with multi-scale convolutional kernel is proposed to exploit discriminative representations from two-dimensional (2D) satellite images.