In recent years, with the extensive application of deep learning in images, the task of remote sensing image change detection has witnessed a significant improvement. Several excellent methods based on Convolutional Neural Networks and emerging transformer-based methods have achieved impressive accuracy. However, Convolutional Neural Network-based approaches have difficulties in capturing long-range dependencies because of their natural limitations in effective receptive field acquisition unless deeper networks are employed, introducing other drawbacks such as an increased number of parameters and loss of shallow information. The transformer-based methods can effectively learn the relationship between different regions, but the computation is inefficient. Thus, in this paper, a multi-scale feature aggregation via transformer (MFATNet) is proposed for remote sensing image change detection. To obtain a more accurate change map after learning the intra-relationships of feature maps at different scales through the transformer, MFATNet aggregates the multi-scale features. Moreover, the Spatial Semantic Tokenizer (SST) is introduced to obtain refined semantic tokens before feeding into the transformer structure to make it focused on learning more crucial pixel relationships. To fuse low-level features (more fine-grained localization information) and high-level features (more accurate semantic information), and to alleviate the localization and semantic gap between high and low features, the Intra- and Inter-class Channel Attention Module (IICAM) are integrated to further determine more convincing change maps. Extensive experiments are conducted on LEVIR-CD, WHU-CD, and DSIFN-CD datasets. Intersection over union (IoU) of 82.42 and F1 score of 90.36, intersection over union (IoU) of 79.08 and F1 score of 88.31, intersection over union (IoU) of 77.98 and F1 score of 87.62, respectively, are achieved. The experimental results achieved promising performance compared to certain previous state-of-the-art change detection methods.

Change detection is an important task in remote sensing imagery which has seen tremendous advances in recent years with the progress of deep neural networks. Although Convolutional Neural Network (CNN) and Transformer-based methods work sufficiently well, the building specificity is usually less considered to contribute to performance improvement. In this letter, we propose a global and specific high-frequency cues guidance network for building change detection, called GSHF-Net. More specifically, a high-frequency cues extraction module is introduced to provide global building-specific edge cues, which consists of a high-frequency cues extraction block and an attention-guidance block. Subsequently, we design a result-aware alignment to mitigate the effect of building slight offset in co-registered images on the introduction of high-frequency information to aid edge recovery. Extensive experiments were performed on LEVIR-CD and WHU-CD, and our GSHF-Net reached state-of-the-art results of F1 90.18 and 89.53, respectively.