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.

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>

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>

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 Change Detection Method of High Resolution Remote Sensing Image Based on D-S Evidence Theory Feature Fusion

IEEE Access ◽  
2021 ◽  
Vol 9 ◽  
pp. 4673-4687
Author(s):  
Jixiang Zhao ◽  
Shanwei Liu ◽  
Jianhua Wan ◽  
Muhammad Yasir ◽  
Huayu Li
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Change Detection ◽  
Detection Method ◽  
Feature Fusion ◽  
Evidence Theory ◽  
Remote Sensing Image

scholarly journals Change Detection in Multispectral Remote Sensing Images with Leader Intelligence PSO and NSCT Feature Fusion

2020 ◽  
Vol 9 (7) ◽  
pp. 462
Author(s):  
Josephina Paul ◽  
B. Uma Shankar ◽  
Balaram Bhattacharyya
Keyword(s):  
Remote Sensing ◽  
Change Detection ◽  
Feature Fusion ◽  
Spectral Band ◽  
Minimum Energy ◽  
Low Frequency ◽  
Remote Sensing Images ◽  
Difference Image ◽  
Difference Images

Change detection (CD) using Remote sensing images have been a challenging problem over the years. Particularly in the unsupervised domain it is even more difficult. A novel automatic change detection technique in the unsupervised framework is proposed to address the real challenges involved in remote sensing change detection. As the accuracy of change map is highly dependent on quality of difference image (DI), a set of Normalized difference images and a complementary set of Normalized Ratio images are fused in the Nonsubsampled Contourlet Transform (NSCT) domain to generate high quality difference images. The NSCT is chosen as it is efficient in suppressing noise by utilizing its unique characteristics such as multidirectionality and shift-invariance that are suitable for change detection. The low frequency sub bands are fused by averaging to combine the complementary information in the two DIs, and, the higher frequency sub bands are merged by minimum energy rule, for preserving the edges and salient features in the image. By employing a novel Particle Swarm Optimization algorithm with Leader Intelligence (LIPSO), change maps are generated from fused sub bands in two different ways: (i) single spectral band, and (ii) combination of spectral bands. In LIPSO, the concept of leader and followers has been modified with intelligent particles performing Lévy flight randomly for better exploration, to achieve global optima. The proposed method achieved an overall accuracy of 99.64%, 98.49% and 97.66% on the three datasets considered, which is very high. The results have been compared with relevant algorithms. The quantitative metrics demonstrate the superiority of the proposed techniques over the other methods and are found to be statistically significant with McNemar’s test. Visual quality of the results also corroborate the superiority of the proposed method.

Sign in / Sign up

Export Citation Format

Share Document