Hyperspectral image change detection using two-branch Unet network with feature fusion

2021 ◽  
Author(s):  
Qiuxia Li ◽  
Tingkui Mu ◽  
Yusen Feng ◽  
Hang Gong ◽  
Feng Han ◽  
...  
Keyword(s):  
Change Detection ◽  
Hyperspectral Image ◽  
Feature Fusion ◽  
Image Change Detection

scholarly journals An Efficient Lightweight Neural Network for Remote Sensing Image Change Detection

2021 ◽  
Vol 13 (24) ◽  
pp. 5152
Author(s):  
Kaiqiang Song ◽  
Fengzhi Cui ◽  
Jie Jiang
Keyword(s):  
Remote Sensing ◽  
Change Detection ◽  
Land Surface ◽  
Network Architecture ◽  
Data Augmentation ◽  
Feature Fusion ◽  
Feature Representation ◽  
Geometric Transformation ◽  
Computation Efficiency ◽  
Image Change Detection

Remote sensing (RS) image change detection (CD) is a critical technique of detecting land surface changes in earth observation. Deep learning (DL)-based approaches have gained popularity and have made remarkable progress in change detection. The recent advances in DL-based methods mainly focus on enhancing the feature representation ability for performance improvement. However, deeper networks incorporated with attention-based or multiscale context-based modules involve a large number of network parameters and require more inference time. In this paper, we first proposed an effective network called 3M-CDNet that requires about 3.12 M parameters for accuracy improvement. Furthermore, a lightweight variant called 1M-CDNet, which only requires about 1.26 M parameters, was proposed for computation efficiency with the limitation of computing power. 3M-CDNet and 1M-CDNet have the same backbone network architecture but different classifiers. Specifically, the application of deformable convolutions (DConv) in the lightweight backbone made the model gain a good geometric transformation modeling capacity for change detection. The two-level feature fusion strategy was applied to improve the feature representation. In addition, the classifier that has a plain design to facilitate the inference speed applied dropout regularization to improve generalization ability. Online data augmentation (DA) was also applied to alleviate overfitting during model training. Extensive experiments have been conducted on several public datasets for performance evaluation. Ablation studies have proved the effectiveness of the core components. Experiment results demonstrate that the proposed networks achieved performance improvements compared with the state-of-the-art methods. Specifically, 3M-CDNet achieved the best F1-score on two datasets, i.e., LEVIR-CD (0.9161) and Season-Varying (0.9749). Compared with existing methods, 1M-CDNet achieved a higher F1-score, i.e., LEVIR-CD (0.9118) and Season-Varying (0.9680). In addition, the runtime of 1M-CDNet is superior to most, which exhibits a better trade-off between accuracy and efficiency.

Remote Sensing Image Change Detection Based on Density Attraction and Multi-Scale and Multi-Feature Fusion

2019 ◽  
Vol 56 (12) ◽  
pp. 121003
Author(s):  
金秋含 Qiuhan Jin ◽  
王阳萍 Yangping Wang ◽  
杨景玉 Jingyu Yang
Keyword(s):  
Remote Sensing ◽  
Change Detection ◽  
Feature Fusion ◽  
Remote Sensing Image ◽  
Multi Scale ◽  
Image Change Detection

scholarly journals 3M-CDNet: A Universal Powerful Benchmark for Remote Sensing Image Change Detection

2021 ◽  
Author(s):  
Kaiqiang Song ◽  
Jie Jiang
Keyword(s):  
Remote Sensing ◽  
Change Detection ◽  
Data Augmentation ◽  
Modular Design ◽  
Feature Fusion ◽  
Feature Representation ◽  
Geometric Transformation ◽  
Generalization Capability ◽  
Online Data ◽  
Image Change Detection

<p><i>Abstract</i>—While deep learning-based methods have gained popularity and have made remarkable progress in remote sensing (RS) image change detection (CD), the limited amount of available data hinders the performance of most supervised methods. The CD networks transferred or derived from other fields can be fronted with a weak generalization capability. Developing a universal benchmark for performance evaluations based on the available datasets is urgent. To address these problems, we proposed a lightweight network, termed 3M-CDNet, which only requires about 3.12 <i>M</i> parameters. The lighter the network, the easier it is to train and alleviate overfitting the limited amount of data, resulting in a better generalization capability. 3M-CDNet has a flexible modular design that achieves performance improvements by incorporating plug-and-play modules. 3M-CDNet gains accuracy improvements in two ways: (1) the application of deformable convolutions (<i>DConv</i>) in the backbone network to gain a good geometric transformation modeling capacity for CD and (2) the application of an effective two-level feature fusion strategy to enhance the feature representation capacity. 3M-CDNet gains a good generalization capacity by incorporating effective “tricks” to alleviate overfitting, in which online data augmentation (<i>Online DA</i>) is applied to increase the diversity of the training samples, and <i>Dropout</i> regularization is applied in the classifier. Extensive ablation studies have proved the effectiveness of the core components. Experiment results suggest that 3M-CDNet outperforms state-of-the-art methods on several optical RS datasets and serves as a new universal benchmark. Specifically, 3M-CDNet achieves the best F1-score, i.e., LEVIR-CD (0.9161), Season-Varying (0.9473), and DSIFN (0.7031). </p>

Sign in / Sign up

Export Citation Format

Share Document