Unsupervised Deep Change Vector Analysis for Multiple-Change Detection in VHR Images

By Sudipan Saha, Francesca Bovolo, Lorenzo Bruzzone. IEEE Transactions on Geoscience and Remote Sensing, 2019

489 citations recorded by the source corpus.

Change detection (CD) in multitemporal images is an important application of remote sensing. Recent technological evolution provided very high spatial resolution (VHR) multitemporal optical satellite images showing high spatial correlation among pixels and requiring an effective modeling of spatial context to accurately capture change information. Here, we propose a novel unsupervised context-sensitive framework—deep change vector analysis (DCVA)—for CD in multitemporal VHR images that exploit convolutional neural network (CNN) features. To have an unsupervised system, DCVA starts from a suboptimal pretrained multilayered CNN for obtaining deep features that can model spatial relationship among neighboring pixels and thus complex objects. An automatic feature selection strategy is employed layerwise to select features emphasizing both high and low prior probability change information. Selected features from multiple layers are combined into a deep feature hypervector providing a multiscale scene representation. The use of the same pretrained CNN for semantic segmentation of single images enables us to obtain coherent multitemporal deep feature hypervectors that can be compared pixelwise to obtain deep change vectors that also model spatial context information. Deep change vectors are analyzed based on their magnitude to identify changed pixels. Then, deep change vectors corresponding to identified changed pixels are binarized to obtain a compressed binary deep change vectors that preserve information about the direction (kind) of change. Changed pixels are analyzed for multiple CD based on the binary features, thus implicitly using the spatial information. Experimental results on multitemporal data sets of Worldview-2, Pleiades, and Quickbird images confirm the effectiveness of the proposed method.

Subfields: Artificial Intelligence, Atmospheric Science, Media Technology

View Paper: https://www.semanticscholar.org/paper/58139ebda484a3b857db2d08f94ad7e8f7cc7686

How researchers see this paper

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

In Artificial Intelligence

1. Lorenzo Bruzzone
2. Yuxing Chen
3. Hongruixuan Chen
4. Le Wang
5. Sicong Liu
6. Rubén Fernández-Beltran
7. Chen Wu
8. Antonio J. Plaza

In Atmospheric Science

1. Lorenzo Bruzzone
2. Yuxing Chen
3. Antonio J. Plaza
4. Bo Du
5. Le Wang
6. Hongruixuan Chen
7. Chen Wu
8. Sicong Liu

In Media Technology

1. Lorenzo Bruzzone
2. Antonio J. Plaza
3. Bo Du
4. Le Wang
5. Yuxing Chen
6. Sicong Liu
7. Chen Wu
8. Hongruixuan Chen

Citations in the corpus

• SemiCDNet: A Semisupervised Convolutional Neural Network for Change Detection in High Resolution Remote-Sensing Images (2020, methodology): “In FB-DLCD methods, deep features are used to generate a DI, then a threshold segmentation method is applied to obtain a final change map (CM) [28], [29].”
• Deep Unsupervised Embedding for Remotely Sensed Images Based on Spatially Augmented Momentum Contrast (2020, background): “applications [1], [2], such as scene classification [3]–[7], object localization [8]–[12], and change detection [13]–[15], due to their prominent capabilities to uncover highly representative features from RS scenes [16].”
• Change Detection in Multisource VHR Images via Deep Siamese Convolutional Multiple-Layers Recurrent Neural Network (2020, methodology): “[26] combined CVA with pretrained deep convolutional neural network (DCNN) and developed deep change vector analysis (DCVA) for VHR image CD. Lyu et al .”
• Deep learning for change detection in remote sensing: a review (2022, methodology): “Direct application means using pre-trained neural networks from other kinds of data to extract deep features (Saha, Bovolo, and Bruzzone 2019; El Amin, Liu, and Wang 2017).”
• Self-Supervised Change Detection in Multiview Remote Sensing Images (2021, methodology): “Recently, deep learning techniques and in particular Convolutional Neural Networks (CNNs) methods [6] have been widely used in this domain.”
• Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges (2020, methodology): “Without fine-tuning, final change maps can be obtained based on the two-period feature maps using change analysis, such as low rank analysis [173], CVA [72], clustering [73,82], and threshold [119,174].”
• Deep Learning-Based Change Detection in Remote Sensing Images: A Review (2022, methodology): “Few studies, such as [149,150,152,153,163], have focused on multiple change maps.”
• Change Detection on Remote Sensing Images Using Dual-Branch Multilevel Intertemporal Network (2023, methodology): “In contrast, image transformation methods, such as principal component analysis (PCA) [7], change vector analysis [8], and tasseled cap transformation [9], resort the spatial mapping to highlight the changed information.”
• ChangeMamba: Remote Sensing Change Detection With Spatiotemporal State Space Model (2024, methodology): “In [37] and [11], unsupervised SCD methods were developed, using pre-trained CNNs or unsupervised learning techniques to extract temporal features, then comparing these features to classify changes.”
• DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation (2022, methodology): “Classical approaches define axiomatic, pixel-based [4–6, 20, 35] algorithms to obtain change whereas many recent approaches are datadriven [7, 12, 33, 47].”