2019IEEE Transactions of Geoscience and Remote Sensing武大變化檢測新模型：DSFA
1.論文地址：Unsupervised Deep Slow Feature Analysis for Change Detection in Multi-Temporal Remote Sensing Images
摘要:Change detection has been a hotspot in remote sensing technology for a long time. With the increasing availability of multi-temporal remote sensing images, numerous change detection algorithms have been proposed. Among these methods, image transformation methods with feature extraction and mapping could effectively highlight the changed information and thus has better change detection performance. However, changes of multi-temporal images are usually complex, existing methods are not effective enough. In recent years, deep network has shown its brilliant performance in many fields including feature extraction and projection. Therefore, in this paper, based on deep network and slow feature analysis (SFA) theory, we proposed a new change detection algorithm for multi-temporal remotes sensing images called Deep Slow Feature Analysis (DSFA). In DSFA model, two symmetric deep networks are utilized for projecting the input data of bi-temporal imagery. Then, the SFA module is deployed to suppress the unchanged components and highlight the changed components of the transformed features. The CVA pre-detection is employed to find unchanged pixels with high confidence as training samples. Finally, the change intensity is calculated with chi-square distance and the changes are determined by threshold algorithms. The experiments are performed on two real-world datasets and a public hyperspectral dataset. The visual comparison and quantitative evaluation have both shown that DSFA could outperform the other state-of-the-art algorithms, including other SFA-based and deep learning methods.
2.開源代碼：https://github.com/rulixiang/DSFANet
3.算法結構
4.代碼運行環境：
tensorflow_gpu==1.7.0
scipy==1.0.0
numpy==1.14.0
matplotlib==2.1.2
tensorflow==1.14.0
5.數據格式樣例地址：dataset
其中：river_before.mat文件爲時相1的圖像轉爲mat文件格式
river_after.mat文件爲時相2的圖像轉爲mat文件格式
change.bmp爲0-255的變化區域二值圖
unchange.bmp爲0-255的未變化區域二值圖
（unchange.bmp就是將change.bmp中的0變成255，255變成0，可以用一句代碼實現）
unchange_bmp = 255 - change_bmp
ind.mat爲0-1的label的mat文件格式
讀取mat文件方法可以參考博客：loadmat
6.運行代碼：
python dsfa.py [-h] [-e EPOCH] [-l LR] [-r REG] [-t TRN] [-i ITER] [-g GPU]
[--area AREA]
#optional arguments:
#-h, --help show this help message and exit
#-e EPOCH, --epoch EPOCH epoches
#-l LR, --lr LR learning rate
#-r REG, --reg REG regularization parameter
#-t TRN, --trn TRN number of training samples
#-i ITER, --iter ITER max iteration
#-g GPU, --gpu GPU GPU ID
#--area AREA datasets
7.結果展示
如有問題，歡迎指正！！！！！