TDSCCNet: twin-depthwise separable convolution connect network for change detection with heterogeneous images

Abstract

The task of change detection (CD) in optical and SAR images is an ever-evolving and demanding subject within the realm of remote sensing (RS). It holds great significance to identify the target areas by using complementary information between the two. Due to the distinct imaging mechanisms employed by optical and SAR sensors, effectively and accurately identifying changing regions can be challenging. To this end, a novel heterogeneous RS images CD network (Twin-Depthwise Separable Convolution Connect Network, TDSCCNet) is proposed in this paper. Image domain transformation is a front-end task, while the back-end employs a single-branch bilayer depthwise separable convolution-connected encoder-decoder to accomplish CD work. Specifically, first, the cycle-consistent adversarial network (CycleGAN) serves to integrate the optical and SAR visual domains, and a consistent feature expression is obtained. Second, the single-branch encoder structure of bilayer depthwise separable convolution is employed to realize change feature extraction. Finally, the multiscale connected decoder reconstructed by change map is utilized to reconstruct the original images and solve the local discontinuities and holes in the binary change map. Multiscale loss is designated for optimizing the global and local effects to alleviate the class imbalance problem. It was tested on four representative datasets from Gloucester, Shuguang Village, Italy and WV-3 datasets with an overall accuracy of 97.36%, 97.01%, 97.62%, and 98.01% respectively. By comparing the existing methods, experimental results confirmed the effectiveness of the proposed method.

KEWORDS:

• Heterogeneous images
• change detection
• image domain transformation
• TDSCCNet
• deep learning

Acknowledgements

The authors express their gratitude to the teams who provided the datasets and algorithms used in this research. We extend our heartfelt thanks to the reviewers and editors for their thorough efforts and valuable insights that significantly improved the quality of this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data used in this study are available by contacting the corresponding author.

Additional information

Funding

This work was Supported by National Natural Science Foundation of China [grant NO. 42361054], Yunnan Fundamental Research Project [grant NO. 202201AT070164], Hunan Provincial Natural Science Foundation of China [grant NO. 2023JJ60561], ‘Xingdian’ talent support program project and Yunnan Province key research and development program [grant NO. 202202AD080010].