STransUNet: A Siamese TransUNet-Based Remote Sensing Image Change Detection Network

Abstract

In modern remote sensing image change detection (CD), convolution Neural Network (CNN), especially U-shaped structure (UNet), has achieved great success due to powerful discriminative ability. However, UNet-based CNN networks usually have limitations in modeling global dependencies due to intrinsic locality of convolution operations. Transformer has recently emerged as an alternative architecture for dense prediction tasks due to global self-attention mechanism. However, due to limitation of hardware resources, pure Transformer methods generally lack the ability to capture global information at a low level. Based on these existing problems, we propose STransUNet, which combines Transformer and UNet architecture. STransUNet can not only capture shallow detail features at an early stage, but also model global context in high-level feature. In addition, we design an efficient feature fusion module named Cross-Enhanced Adaptive Fusion(CEAF). Our model mainly consists of three parts: encoder, fusion module and decoder. The decoder is a CNN-Transformer hybrid structure. CNN extracts multi-level feature information. Transformer encodes tokenized sequence to capture global context. CEAF module cross-enhances and adaptively fuses bi-temporal features to enhance feature representation. In decoding stage, we introduce a Cascaded Upsampling decoder(CUP). CUP progressively aggregates lowlevel CNN features and high-level Transformer features to full resolution. On four public CD datasets, our STransUNet achieves better CD results than six state-of-the-art algorithms.

MDFENet: A Multiscale Difference Feature Enhancement Network for Remote Sensing Change Detection

“…Hybrid-TransCD [62] proposed a hybrid multi-scale transformer structure, which can model the mixed-scale attention representation of each image. STransUNet [66] combined transformer and UNet architectures, which can capture shallow detail features and model global context in high-level features. In order to capture the spatial and channel information of feature maps, MTCNet [67] divides CBAM into a SAM and a CAM, which are applied to the front-end and back-end of the multi-scale transformer, respectively.…”

A hybrid U‐shaped and transformer network for change detection in high‐resolution remote sensing images

“…To simplify the transformer block, Lin et al [44] presented the swin transformer framework by incorporating hierarchical, locality, and shift invariance priors, which achieves better performance in various tasks. Inspired by transformer, some CD methods based on transformer and CNN are proposed [45][46][47][48][49][50]. Shi et al [51] designed the bi-temporal image transformer (BIT), a method based on transformer to address CD challenges.…”

“…[44] presented the swin transformer framework by incorporating hierarchical, locality, and shift invariance priors, which achieves better performance in various tasks. Inspired by transformer, some CD methods based on transformer and CNN are proposed [45–50]. Shi et al.…”

A Multi-Task Consistency Enhancement Network for Semantic Change Detection in HR Remote Sensing Images and Application of Non-Agriculturalization

“…In contrast, Transformer possesses prominent global modeling capability, and it is independent of the constraints of a traditional convolutional receptive field, thus making it more versatile and flexible [31]. However, a drawback of Transformer lies in its dependence on a greater number of deep feature sequences to indicate its performance, such that local fine granularity and the large computational parameter become difficult to perceive [32,33]. Thus, combining the strengths of CNN and Transformer in local and global modeling has effectively become an advanced design approach.…”