# HybridSN

**Repository Path**: xyt556/HybridSN

## Basic Information

- **Project Name**: HybridSN
- **Description**: A keras based implementation of Hybrid-Spectral-Net as in our paper "HybridSN: Exploring 3D-2D CNN Feature Hierarchy for Hyperspectral Image Classification".
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2020-08-12
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Hybrid-Spectral-Net for Hyperspectral Image Classification.
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/hybridsn-exploring-3d-2d-cnn-feature/hyperspectral-image-classification-on-indian)](https://paperswithcode.com/sota/hyperspectral-image-classification-on-indian?p=hybridsn-exploring-3d-2d-cnn-feature)
## Description

The  HybridSN  is  spectral-spatial  3D-CNN  followed  by spatial 2D-CNN. The 3D-CNN facilitates the joint spatial-spectral feature  representation  from  a  stack  of  spectral  bands.  The  2D-CNN  on  top  of  the  3D-CNN  further  learns  more  abstract  level spatial  representation. 

## Model

<img src="figure/HSI-RN.jpg"/>

Fig: Proposed HybridSpectralNet (HybridSN) Model with 3D and 2D convolutions for hyperspectral image (HSI) classification.

## Prerequisites

- [Anaconda 2.7](https://www.anaconda.com/download/#linux)
- [Tensorflow 1.3](https://github.com/tensorflow/tensorflow/tree/r1.3)
- [Keras 2.0](https://github.com/fchollet/keras)

## Results

### Indian Pines (IP) dataset

<img src="figure/IP-FC.jpg" width="200" height="200"/> <img src="figure/IP-GT.jpg" width="200" height="200"/> <img src="figure/IP-Pr.jpg" width="200" height="200"/> <img src="figure/IP_legend.jpg" width="250" height="150"/>

Fig.2  The IN dataset classification result (Overall Accuracy 99.81%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map. (d) Class legend. 

### University of Pavia (UP) dataset

<img src="figure/UP-FC.jpg"/> <img src="figure/UP-GT.jpg"/> <img src="figure/UP-Pr.jpg"/> <img src="figure/UP_legend.jpg" width="200" height="100"/>

Fig.3  The UP dataset classification result (Overall Accuracy 99.99%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map. (d) Class legend.

### Salinas Scene (SS) dataset

<img src="figure/SA-FC.jpg"/> <img src="figure/SA-GT.jpg"/> <img src="figure/SA-Pr.jpg"/> <img src="figure/SA_legend.jpg" width="300" height="150"/>

Fig.4  The UP dataset classification result (Overall Accuracy 100%) of Hybrid-SN using 30% samples for training. (a) False color image. (b) Ground truth labels. (c) Classification map.

#### Detailed results can be found in the [Supplementary Material](supplementary-material.pdf)

## Citation

If you use this code in your research, we would appreciate a citation to the original [paper](paper.pdf):

	@article{roy2019hybridsn,
            title={HybridSN: Exploring 3D-2D CNN Feature Hierarchy for Hyperspectral Image Classification},
            author={Roy, Swalpa Kumar and Krishna, Gopal and Dubey, Shiv Ram and Chaudhuri, Bidyut B},
            journal={IEEE Geoscience and Remote Sensing Letters},
            year={2019}
            }


## Acknowledgement

Part of this code is from a implementation of Classification of HSI using CNN by [Konstantinos Fokeas](https://github.com/KonstantinosF/Classification-of-Hyperspectral-Image).

## License

Copyright (c) 2019 Gopal Krishna. Released under the MIT License. See [LICENSE](LICENSE) for details.