SiUNet3+-CD: a full-scale connected Siamese network for change detection of VHR images

Zhao, Bo; Tang, Panpan; Luo, Xiaoyan; Li, Linze; Bai, Shi

doi:10.1080/22797254.2022.2047795

Cited by 7 publications

(25 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we use 𝛼 = 0.25 and 𝛾 = 2 through trial and error. Dice loss, which is used in the medical image segmentation field, is applied to address the imbalanced dataset problem [45]. The equation of Dice loss is as follows:…”

Section: B Loss Function For Training the Proposed Networkmentioning

confidence: 99%

“…In addition, the proposed model was processed through the concatenation process of the layers created by Siamese architecture. However, Zhao et al [45] proposed SiUNet3+-CD based on the UNet3+ for change detection in satellite images. They have similar network structures, compared to the proposed model in that they use UNet3+ as the backbone for change detection.…”

Section: F Ablation Study On the Levir-cd Datasetmentioning

confidence: 99%

See 1 more Smart Citation

SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images

Seong

et al. 2022

IEEE Access

View full text Add to dashboard Cite

With the development of various optical sensors, change detection is one of the most actively researched areas in remotely sensed imagery with high spatial resolution. In particular, deep learning-based change detection techniques are very important for use in various fields, such as land monitoring and disaster analysis, because they can show superior performance compared to traditional unsupervised and supervised change detection methods. In this manuscript, we propose a Siamese-attentive UNet3+ for change detection (SAUNet3+CD) of multitemporal imagery with high spatial resolution. The existing UNet3+ was modified to a Siamese-based architecture, and a spatial and channel attention module was added to detect various changed areas. The proposed model was trained to effectively detect both building growth and decay through the data augmentation of open datasets and a hybrid loss function. In experiments using two open datasets, the proposed deep learning model effectively detected changed areas in multitemporal images better than various methods, such as existing Siamese-based networks and a network for semantic segmentation.

show abstract

Section: B Loss Function For Training the Proposed Networkmentioning

confidence: 99%

Section: F Ablation Study On the Levir-cd Datasetmentioning

confidence: 99%

SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images

Seong

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, although the fine contextual information and complex spatial characteristics that high-resolution images convey offer rich spatial details for land use/cover change analysis, they face several serious challenges too [4]. Difficulties arise from the enhanced intra-class variabilities [5], the spatial displacement due to the parallax distortion of ground objects (especially for high-rise buildings) [6], the confused spectral features of different objects, complex scenarios, illumination, camera motion, shadows, misregistration error, and etc.…”

Section: Introductionmentioning

confidence: 99%

“…Due to the great advantages in deep feature representation and non-linear problem modeling [5], deep learning technology has opened up new opportunities for CD tasks to address the problems above-mentioned. For example, the CNN (i.e., Convolutional Neural Networks) architecture itself could achieve some degree of shift, scale, and distortion invariance [7].…”

Section: Introductionmentioning

confidence: 99%

“…For example, the CNN (i.e., Convolutional Neural Networks) architecture itself could achieve some degree of shift, scale, and distortion invariance [7]. Over the past years, enormous efforts have been made to use deep learning-based methods for CD using high-resolution RS images, readers are referred to [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], and [20]. A wide variety of research has indicated the superiority of these methods over the traditional machine learning or unsupervised methods, e.g., Support Vector Machine, Random Forest, Slow Feature Analysis [21], Robust Change Vector Analysis [22], and so on.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3M-CDNet-V2: An Efficient Medium-Weight Neural Network for Remote Sensing Image Change Detection

Zhao¹,

Tang²,

Luo³

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

Remote sensing-based change detection (CD) is a critical technique of detecting land surface changes in earth observation. Inspired by recent success of a lightweight CD network -3M-CDNet, we implemented 9 meaningful modifications to it, for example, the incorporation of MHSA (Multi-Head Self-Attention). This elaborately designed model is termed as 3M-CDNet-V2. Its effectiveness and advantages were demonstrated on three engineering CD datasets, and experimental results indicated that:(1) relative to other state-of-the-art algorithms, the proposed model obtained very competitive or slight better performance on both visual comparison and quantitative metrics evaluation. (2) By applying a novel transfer learning strategy, 3M-CDNet-V2 could perform well on the small dataset. (3) The incorporation of MHSA brings a substantial accuracy improvement while moderately increasing the computational complexity. (4) The late fusion framework and deep supervision contribute most to the performance gain of 3M-CDNet-V2, while the introduction of low-level features to the classifier by skip connection could guide the model to focus on detailed spatial information such as small changes, narrow shaped objects, or accurate boundaries. We hope that our 3M-CDNet-V2 model helps in improving the understanding of network architecture design for CD.

show abstract

Recognition of multivariate geochemical anomalies using a geologically-constrained variational autoencoder network with spectrum separable module – A case study in Shangluo District, China

Zhao¹,

Zhang

Tang³

et al. 2023

Applied Geochemistry

View full text Add to dashboard Cite

SiUNet3+-CD: a full-scale connected Siamese network for change detection of VHR images

Cited by 7 publications

References 35 publications

SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images

SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images

3M-CDNet-V2: An Efficient Medium-Weight Neural Network for Remote Sensing Image Change Detection

Recognition of multivariate geochemical anomalies using a geologically-constrained variational autoencoder network with spectrum separable module – A case study in Shangluo District, China

Contact Info

Product

Resources

About