Change Detection Based on Deep Siamese Convolutional Network for Optical Aerial Images

By Yang Zhan, Kun Fu, Menglong Yan, Xian Sun, Hongqi Wang, Xiaosong Qiu. 2017

538 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Atmospheric Science, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Analytical Chemistry

In Atmospheric Science

In Media Technology

Citations in the corpus

Recognition in Terra Incognita (2018, background): “[38] and [39] consider low frame rate change detection in aerial images, but in these cases there are often very few examples per location.”
A Spatial-Temporal Attention-Based Method and a New Dataset for Remote Sensing Image Change Detection (2020, methodology): “The embedding space can be learned by deep Siamese fully convolutional networks (FCN) [27,28], which contains two identical networks sharing the same weight, each independently generating the feature maps for each temporal image.”
Fully Convolutional Siamese Networks for Change Detection (2018, methodology): “…began to be used in conjunction with simple machine learning techniques [6]; recently, more elaborate machine learning techniques (deep learning) are dominating most problems in the image analysis ﬁeld, and this evolution is slowly reaching the problem of change detection [7, 8, 9, 10, 11, 3, 12].”
DASNet: Dual Attentive Fully Convolutional Siamese Networks for Change Detection in High-Resolution Satellite Images (2020, methodology): “SCCN [23] uses a deep symmetrical network to study changes in remote sensing images, and DSCN [24] uses two branch networks that share weights for feature extraction and uses the features that are obtained by the last layer of the two branches for threshold segmentation to obtain a binary change map”
Change Detection in Multisource VHR Images via Deep Siamese Convolutional Multiple-Layers Recurrent Neural Network (2020, methodology): “For the purpose of demonstrating the superiority of SiamCRNN, eight representative methods are utilized for comparison: direct subtraction (DS), postclassiﬁcation comparison (PCC), MAD [14], IRMAD [15], S-MAD [14], SVM, early fusion (EF)-DCNN, and DSCN [25].”
Unsupervised Deep Change Vector Analysis for Multiple-Change Detection in VHR Images (2019, methodology): “[34] proposed a supervised CD method for optical aerial images based on the deep Siamese network.”
Changer: Feature Interaction is What You Need for Change Detection (2022, background): “To achieve this goal, metric learning-related losses, such as contrast loss [15] are leveraged to pull unchanged pairs together and push changed pairs apart [5,45].”
Deep learning for change detection in remote sensing: a review (2022, methodology): “…four basic transformation operations including rotation (Zhan et al. 2017; Nemoto et al. 2017; Wang et al. 2018a; Zhu et al. 2018), flip (Zhan et al. 2017), mirror (Zhu et al. 2018), and cropping (Nemoto et al. 2017; Zhan et al. 2017) are directly used to generate augmented training samples.”
SwinSUNet: Pure Transformer Network for Remote Sensing Image Change Detection (2022, background): “Zhan et al. [18] proposed weighted contrastive loss function which increased the weight of the changed pixels in the loss calculation.”
DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation (2022, methodology): “Most methods focus on binary change and are usually limited to two distinct observations in time (bitem-poral) [4–7, 20, 35 , 47 ].”
Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networks (2018, background): “Moreover, the recently proposed deep learning algorithms for change detection have mostly been designed to generate a difference image which is manually thresholded [6].”
Change Detection on Remote Sensing Images Using Dual-Branch Multilevel Intertemporal Network (2023, methodology): “Incorporating certain clustering method to strengthen the discriminative characteristics of CNN is explored in [24] and [25].”
Multitask learning for large-scale semantic change detection (2018, methodology): “CNN architectures have also been proposed to perform supervised change detection (Zhan et al., 2017; Chen et al., 2018b).”
End-to-End Change Detection for High Resolution Satellite Images Using Improved UNet++ (2019, methodology): “Thus, Siamese CNN architectures with weighted contrastive loss [31] and improved triplet loss [32] were exploited to learn discriminative deep features between changed and unchanged pixels, then DIs were generated based on the Euclidean distances of deep features, and finally CM could be obtained by”
Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges (2020, background): “Siamese structure [22,68,94,117,155,156] and the pseudo-Siamese structure [79,109,157,158].”
Self-Supervised Change Detection in Multiview Remote Sensing Images (2021, background): “achieving state-of-the-art results in a supervised way [7].”
Building Change Detection for Remote Sensing Images Using a Dual-Task Constrained Deep Siamese Convolutional Network Model (2019, methodology): “[4] used a deep Siamese convolutional network to extract similar feature maps and then clustered them with a k-nearest neighbor (KNN) classifier to obtain the changed area.”
Optical Remote Sensing Image Change Detection Based on Attention Mechanism and Image Difference (2021, background): “The second type is the double-branch networks with shared weights [36], [37] (i.e., the Siamese network).”