Remote sensing Change Detection (CD) involves identifying changing regions of interest in bi-temporal remote sensing images. CD technology has rapidly developed in recent years through the powerful learning ability of Convolutional Neural Networks (CNN), affording complex feature extraction. However, the local receptive fields in the CNN limit modeling long-range contextual relationships in semantic changes. Therefore, this work explores the great potential of Siamese Transformers in CD tasks and proposes a general CD model entitled STCD that relies on Swin Transformers. In the encoding process, pure Transformers without CNN are used to model the long-range context of semantic tokens, reducing computational overhead and improving model efficiency compared to current methods. During the decoding process, the 3D convolution block obtains the changing features in the time series and generates the predicted change map in the deconvolution layer with axial attention. Extensive experiments on three binary CD datasets and one semantic CD dataset demonstrate that the proposed STCD model outperforms several popular benchmark methods considering performance and the required parameters. Among the STCD variants, the F1-Score of the Base-STCD on the three binary CD datasets LEVIR, DSIFN, and SVCD reached 89.85%, 54.72%, and 93.75%, respectively, and the mF1-Score and mIoU on the semantic CD dataset SECOND were 75.60% and 66.19%.

The multiclass change detection task aims to segment an area of land change and to identify the type of ground change exploiting bi-temporal remote-sensing images. In recent years, this technology has become an important means of analysing remote-sensing data and has shown great potential in accurately understanding changes on the earth’s surface. However, the existing methods are mainly for binary change detection tasks and cannot detect the specific information of the change category. Hence, in this paper, we propose an end-to-end densely connected network, entitled Y-Net, employing a Siamese structure for multiclass change detection. The proposed network uses a dual-stream DenseNet to extract bi-temporal change features during the encoding stage and introduces the attention fusion mechanism for the decoding stage to enhance the attention to changing features. Add the dice loss to reduce the impact of sample imbalance on multiclass change detection. The suggested network utilizes a deep supervision strategy for each decoding layer to achieve change graph reconstruction of different levels. Y-Net is evaluated on the challenging multiclass change detection data set SECOND, and the ablation study verified the effectiveness of the attention fusion module and dice loss for multiclass change detection, considering both visual interpretation and quantitative evaluation, confirming that the proposed method is superior to current state-of-the-art change detection methods.

The change detection (CD) of heterogeneous remote sensing images is an important but challenging task. The difficulty is to obtain the change information by directly comparing the different statistical characteristics of the images acquired by different sensors. This paper proposes an unsupervised method for heterogeneous image CD based on an image domain transfer network. First, an attention mechanism is added to the Cycle-generative adversarial networks (Cycle-GANs) to obtain a more consistent feature expression by transferring bi-temporal heterogeneous images to the common domain. The Euclidean distance of the corresponding pixels is calculated in the common domain to form a difference map, and a threshold algorithm is applied to get a rough change map. Finally, the proposed adaptive Discrete Cosine Transform (DCT) algorithm reduces the noise introduced by false detection, and the final change map is obtained. The proposed method is verified on three real heterogeneous CD datasets and compared with the current state-of-the-art methods. The results show that the proposed method is accurate and robust for performing heterogeneous CD tasks.