IRX-1D: A Simple Deep Learning Architecture for Remote Sensing Classifications

TL;DR

This paper introduces IRX-1D, a simple deep learning model combining elements of Inception, ResNet, and Xception, which improves classification accuracy on remote sensing datasets, especially with limited training samples.

Contribution

The paper presents a novel deep learning architecture, IRX-1D, that outperforms existing models in remote sensing classification tasks with small training datasets.

Findings

IRX-1D achieves higher accuracy than Bayesian optimized 2D-CNN.

Comparable or better performance than nine existing deep learning methods.

Limited training samples lead to different land cover classifications compared to large-sample models.

Abstract

We proposes a simple deep learning architecture combining elements of Inception, ResNet and Xception networks. Four new datasets were used for classification with both small and large training samples. Results in terms of classification accuracy suggests improved performance by proposed architecture in comparison to Bayesian optimised 2D-CNN with small training samples. Comparison of results using small training sample with Indiana Pines hyperspectral dataset suggests comparable or better performance by proposed architecture than nine reported works using different deep learning architectures. In spite of achieving high classification accuracy with limited training samples, comparison of classified image suggests different land cover classes are assigned to same area when compared with the classified image provided by the model trained using large training samples with all datasets.