MTCNet: Multitask consistency network with single temporal supervision for semi-supervised building change detection

Shu, Qidi; Pan, Jun; Zhang, Zhuoer; Wang, Mi

doi:10.1016/j.jag.2022.103110

Cited by 16 publications

(11 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 2 lists the quantitative change detection results obtained on the test set from training the models on limited fractions of the labeled training set, i.e., 40% (n = 12), 20 % (n = 6), and 10% (n = 3). The last column of Table 2 lists the results obtained from training the models on the entire labeled training set (i.e., 100%), even though it is generally assumed in semi-supervised learning that the size of the unlabeled dataset is considerably larger than that of the labeled dataset (e.g., [43,45]). However, this column was added to test whether the proposed method manages to perform on par with the supervised method under no label scarcity.…”

Section: Change Detection Resultsmentioning

confidence: 99%

“…Two accuracy metrics were used for the quantitative assessment of predicted changes: F1 score and intersection over union (IoU). The combination of the F1 score and IoU is commonly used for performance assessments in change detection studies, e.g., [45]. Formulas for the metrics are given in Equations 5 and 6, where TP, FP, and FN represent the number of true positive, false positive, and false negative pixels, respectively.…”

Section: Accuracy Metricsmentioning

confidence: 99%

“…First, consistency regularization, following the underlying idea that perturbations of a sample should not significantly change the model output [41,42]; and, second, entropy minimization, which encourages more confident predictions on unlabeled data. Several recent papers applied the former technique, consistency regularization, to urban change detection problems using multi-task Siamese networks [43][44][45]. In particular, Bandara and Patel [43] first used an encoder with shared weights to extract features from unlabeled bitemporal VHR images.…”

Section: Introductionmentioning

confidence: 99%

“…On the other hand, a more recent work aimed to improve change detection by incorporating additional building labels into network training since building labels are less costly to obtain than change labels. To leverage additional building labels, Shu et al [45] proposed a network that encourages consistency between the semantic segmentation of the pre-change image and a building prediction for the pre-change image derived from the change decoder features and the features of the post-change image produced by the semantic decoder. However, despite the fact that these works address the limited availability of labels for urban change detection, up to now, research on semi-supervised change detection has been limited to unimodal EO data from optical sensors.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Semi-Supervised Urban Change Detection Using Multi-Modal Sentinel-1 SAR and Sentinel-2 MSI Data

Hafner,

Ban,

Nascetti

2023

Remote Sensing

View full text Add to dashboard Cite

Urbanization is progressing at an unprecedented rate in many places around the world. The Sentinel-1 synthetic aperture radar (SAR) and Sentinel-2 MultiSpectral Instrument (MSI) missions, combined with deep learning, offer new opportunities to accurately monitor urbanization at a global scale. Although the joint use of SAR and optical data has recently been investigated for urban change detection, existing data fusion methods rely heavily on the availability of sufficient training labels. Meanwhile, change detection methods addressing label scarcity are typically designed for single-sensor optical data. To overcome these limitations, we propose a semi-supervised urban change detection method that exploits unlabeled Sentinel-1 SAR and Sentinel-2 MSI data. Using bitemporal SAR and optical image pairs as inputs, the proposed multi-modal Siamese network predicts urban changes and performs built-up area segmentation for both timestamps. Additionally, we introduce a consistency loss, which penalizes inconsistent built-up area segmentation across sensor modalities on unlabeled data, leading to more robust features. To demonstrate the effectiveness of the proposed method, the SpaceNet 7 dataset, comprising multi-temporal building annotations from rapidly urbanizing areas across the globe, was enriched with Sentinel-1 SAR and Sentinel-2 MSI data. Subsequently, network performance was analyzed under label-scarce conditions by training the network on different fractions of the labeled training set. The proposed method achieved an F1 score of 0.555 when using all available training labels, and produced reasonable change detection results (F1 score of 0.491) even with as little as 10% of the labeled training data. In contrast, multi-modal supervised methods and semi-supervised methods using optical data failed to exceed an F1 score of 0.402 under this condition. Code and data are made publicly available.

show abstract

Section: Change Detection Resultsmentioning

confidence: 99%

Section: Accuracy Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Semi-Supervised Urban Change Detection Using Multi-Modal Sentinel-1 SAR and Sentinel-2 MSI Data

Hafner,

Ban,

Nascetti

2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Remote Sens. 2024, 16, x FOR PEER REVIEW 16 of 36 • Model Perturbation Space [128,129]. This approach involves altering the model itself to create pseudo-labels for unlabeled samples using different models and then supervising them mutually.…”

mentioning

confidence: 99%

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

Wang,

Zhang,

Gao

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Change detection (CD) in remote sensing (RS) imagery is a pivotal method for detecting changes in the Earth’s surface, finding wide applications in urban planning, disaster management, and national security. Recently, deep learning (DL) has experienced explosive growth and, with its superior capabilities in feature learning and pattern recognition, it has introduced innovative approaches to CD. This review explores the latest techniques, applications, and challenges in DL-based CD, examining them through the lens of various learning paradigms, including fully supervised, semi-supervised, weakly supervised, and unsupervised. Initially, the review introduces the basic network architectures for CD methods using DL. Then, it provides a comprehensive analysis of CD methods under different learning paradigms, summarizing commonly used frameworks. Additionally, an overview of publicly available datasets for CD is offered. Finally, the review addresses the opportunities and challenges in the field, including: (a) incomplete supervised CD, encompassing semi-supervised and weakly supervised methods, which is still in its infancy and requires further in-depth investigation; (b) the potential of self-supervised learning, offering significant opportunities for Few-shot and One-shot Learning of CD; (c) the development of Foundation Models, with their multi-task adaptability, providing new perspectives and tools for CD; and (d) the expansion of data sources, presenting both opportunities and challenges for multimodal CD. These areas suggest promising directions for future research in CD. In conclusion, this review aims to assist researchers in gaining a comprehensive understanding of the CD field.

show abstract

MTSCD-Net: A network based on multi-task learning for semantic change detection of bitemporal remote sensing images

Cui

Jiang

2023

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

MTCNet: Multitask consistency network with single temporal supervision for semi-supervised building change detection

Cited by 16 publications

References 31 publications

Semi-Supervised Urban Change Detection Using Multi-Modal Sentinel-1 SAR and Sentinel-2 MSI Data

Semi-Supervised Urban Change Detection Using Multi-Modal Sentinel-1 SAR and Sentinel-2 MSI Data

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

MTSCD-Net: A network based on multi-task learning for semantic change detection of bitemporal remote sensing images

Contact Info

Product

Resources

About