Cascaded attention-induced difference representation learning for multispectral change detection

Zhang, Wuxia; Zhang, Qinyu; Ning, Hailong; Lu, Xiaoqiang

doi:10.1016/j.jag.2023.103366

Cited by 4 publications

(1 citation statement)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the time-domain transformer is used to extract the useful change features between the current image pair, and the detection result is obtained through the multi-layer perceptron. Zhang et al [46] proposed a cascaded attention-induced differential representation learning method for multispectral change detection, which explores the correlation of features extracted from bitemporal images to obtain more discriminative features, and finally detects the discriminative features and obtains the final detection map.…”

Section: Introduction 1backgroundmentioning

confidence: 99%

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Yin,

Ma,

Weng

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Recently, deep learning-based change detection methods for bitemporal remote sensing images have achieved promising results based on fully convolutional neural networks. However, due to the inherent characteristics of convolutional neural networks, if the previous block fails to correctly segment the entire target, erroneous predictions might accumulate in the subsequent blocks, leading to incomplete change detection results in terms of structure. To address this issue, we propose a bitemporal remote sensing image change detection network based on a Siamese-attention feedback architecture, referred to as SAFNet. First, we propose a global semantic module (GSM) on the encoder network, aiming to generate a low-resolution semantic change map to capture the changed objects. Second, we introduce a temporal interaction module (TIM), which is built through each encoding and decoding block, using the feature feedback between two temporal blocks to enhance the network’s perception ability of the entire changed target. Finally, we propose two auxiliary modules—the change feature extraction module (CFEM) and the feature refinement module (FRM)—which are further used to learn the fine boundaries of the changed target. The deep model we propose produced satisfying results in dual-temporal remote sensing image change detection. Extensive experiments on two remote sensing image change detection datasets demonstrate that the SAFNet algorithm exhibits state-of-the-art performance.

show abstract

Section: Introduction 1backgroundmentioning

confidence: 99%

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Yin,

Ma,

Weng

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

Multiview Hypergraph Fusion Network for Change Detection in High-Resolution Remote Sensing Images

Zhao,

Zhang,

Zhang

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Currently, convolution neural networks (CNNs) and transformers have been the dominant paradigms for change detection (CD) thanks to their powerful local and global feature extraction capabilities. However, with the improvement of resolution, spatial, spectral, and temporal relationships among objects in remote sensing images are becoming more complicated, and cannot be modeled efficiently by existing methods. To capture the high-order complex relationships in images, we propose a multiview hypergraph fusion network (MVHFNet) for CD, in which the high-order relationships along spatial, spectral, and temporal views are extracted by hypergraph learning. Specifically, this network is composed of three branches, including the spectral hypergraph learning branch (SpeHGL), the spatial hypergraph learning branch (SpaHGL) branch, and the temporal hypergraph learning branch (TemHGL). In these branches, multiview features are extracted by different attention modules, and hypergraph learning consisting of hypergraph construction and hypergraph convolution is imposed on these features to model the high-order relationships. Then, to integrate the multi-view features from different branches, a multi-view feature fusion (MVF) module is designed, in which the multi-view features are fused and condensed for the following prediction. Finally, the change map is produced by a prediction head. We conduct extensive experiments on three datasets, such as LEVIR-CD, SYSU-CD, and CLCD. The experimental results demonstrate that the proposed MVHFNet achieves better CD performance compared to some state-of-the-art methods.

show abstract

DGMA²-Net: A Difference-Guided Multiscale Aggregation Attention Network for Remote Sensing Change Detection

Ying,

Tan,

Zhai

et al. 2024

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Cascaded attention-induced difference representation learning for multispectral change detection

Cited by 4 publications

References 43 publications

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Multiview Hypergraph Fusion Network for Change Detection in High-Resolution Remote Sensing Images

DGMA²-Net: A Difference-Guided Multiscale Aggregation Attention Network for Remote Sensing Change Detection

Contact Info

Product

Resources

About

Cascaded attention-induced difference representation learning for multispectral change detection

Cited by 4 publications

References 43 publications

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Bitemporal Remote Sensing Image Change Detection Network Based on Siamese-Attention Feedback Architecture

Multiview Hypergraph Fusion Network for Change Detection in High-Resolution Remote Sensing Images

DGMA2-Net: A Difference-Guided Multiscale Aggregation Attention Network for Remote Sensing Change Detection

Contact Info

Product

Resources

About

DGMA²-Net: A Difference-Guided Multiscale Aggregation Attention Network for Remote Sensing Change Detection