acda's Introduction

Hyperspectral anomaly change detection based on autoencoder

Pytorch implementation of JSTARS paper "Hyperspectral anomaly change detection based on autoencoder".

Paper

Please cite our paper if you find it useful for your research.

@ARTICLE{9380336, author={Hu, Meiqi and Wu, Chen and Zhang, Liangpei and Du, Bo}, journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing}, title={Hyperspectral Anomaly Change Detection Based on Autoencoder}, year={2021}, volume={14}, number={}, pages={3750-3762}, doi={10.1109/JSTARS.2021.3066508}}

Installation

Install Pytorch 1.10.2 with Python 3.6

Dataset

Download the [dataset of Viareggio 2013] 链接：https://pan.baidu.com/s/1x_M0nRqV-jmugIB6MltmXQ 提取码：ogum

[Dataset]: "Viareggio 2013" with de-striping, noise-whitening and spectrally binning

img_data.mat:

img_1(D1F12H1); img_2(D1F12H2); img_3(D2F22H2)

pretrain_samples:

un_idx_train1,un_idx_valid1,un_idx_train2,un_idx_valid2; [acquired from the pre-detection result of USFA, Wu C, Zhang L, Du B. Hyperspectral anomaly change detection with slow feature analysis[J]. Neurocomputing, 2015, 151: 175-187.]

groundtruth_samples:

un_idx_train1,un_idx_valid1,un_idx_train2,un_idx_valid2;

random_samples: un_idx_train1,un_idx_valid1,un_idx_train2,un_idx_valid2;

Usage

maincode.py

🌷Homepage🌷
🔴Google web 🔴
🌏ResearchGate 🌍

acda's People

Contributors

Stargazers

Watchers

acda's Issues

师姐，请教一个问题

您是如何理解异常变化检测和变化检测二者的本质区别的呢？

Viareggio 2013 img_data数据中的img_2和img_2_RE的关系

你好，谢谢你的分享。
请问 ProcessDataForACDA 中 img_data 里的img_2和img_2_RE的关系是什么呢？同理img_3和img_3_RE。

Can you share the code of the comparative experiment

I studied this direction for the first time and wanted to learn better, but the code is very scarce. Can you share the code of the comparative experiment？

Recommend Projects