Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection
Stripe-like space target detection (SSTD) plays a crucial role in advancing space situational awareness, enabling missions like satellite navigation and debris monitoring. Existing unsupervised methods often falter in low signal-to-noise ratio (SNR) conditions, while fully supervised approaches requ...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/8/1341 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849713807767633920 |
|---|---|
| author | Zijian Zhu Ali Zia Xuesong Li Bingbing Dan Yuebo Ma Hongfeng Long Kaili Lu Enhai Liu Rujin Zhao |
| author_facet | Zijian Zhu Ali Zia Xuesong Li Bingbing Dan Yuebo Ma Hongfeng Long Kaili Lu Enhai Liu Rujin Zhao |
| author_sort | Zijian Zhu |
| collection | DOAJ |
| description | Stripe-like space target detection (SSTD) plays a crucial role in advancing space situational awareness, enabling missions like satellite navigation and debris monitoring. Existing unsupervised methods often falter in low signal-to-noise ratio (SNR) conditions, while fully supervised approaches require extensive and labor-intensive pixel-level annotations. To address these limitations, this paper introduces MRSA-Net, a novel encoder-decoder network specifically designed for SSTD. MRSA-Net incorporates multi-receptive field processing and multi-level feature fusion to effectively extract features of variable and low-SNR stripe-like targets. Building upon this, we propose the Collaborative Static-Dynamic Teaching (CSDT) architecture, a semi-supervised learning architecture that reduces reliance on labeled data by leveraging both static and dynamic teacher models. The framework uses the straight-line prior of stripe-like targets to customize linearity and presents an innovative Adaptive Pseudo-Labeling (APL) strategy, dynamically selecting high-quality pseudo-labels to enhance the student model’s learning process. Extensive experiments on AstroStripeSet and other real-world datasets demonstrate that the CSDT framework achieves state-of-the-art performance in SSTD. Using just 1/16 of the labeled data, CSDT outperforms the second-best Interactive Self-Training Mean Teacher (ISMT) method by 2.64% in mean Intersection over Union (mIoU) and 4.5% in detection rate (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>P</mi><mi>d</mi></msub></semantics></math></inline-formula>), while exhibiting strong generalization in unseen scenarios. This work marks the first application of semi-supervised learning techniques to SSTD, offering a flexible and scalable solution for challenging space imaging tasks. |
| format | Article |
| id | doaj-art-6970a4ae8ed741e09d2ec0ba72d0f775 |
| institution | DOAJ |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-6970a4ae8ed741e09d2ec0ba72d0f7752025-08-20T03:13:51ZengMDPI AGRemote Sensing2072-42922025-04-01178134110.3390/rs17081341Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target DetectionZijian Zhu0Ali Zia1Xuesong Li2Bingbing Dan3Yuebo Ma4Hongfeng Long5Kaili Lu6Enhai Liu7Rujin Zhao8National Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaCollege of Science, Australian National University, Canberra 2601, AustraliaCollege of Science, Australian National University, Canberra 2601, AustraliaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaNational Key Laboratory of Optical Field Manipulation Science and Technology, Chinese Academy of Sciences, Chengdu 610209, ChinaStripe-like space target detection (SSTD) plays a crucial role in advancing space situational awareness, enabling missions like satellite navigation and debris monitoring. Existing unsupervised methods often falter in low signal-to-noise ratio (SNR) conditions, while fully supervised approaches require extensive and labor-intensive pixel-level annotations. To address these limitations, this paper introduces MRSA-Net, a novel encoder-decoder network specifically designed for SSTD. MRSA-Net incorporates multi-receptive field processing and multi-level feature fusion to effectively extract features of variable and low-SNR stripe-like targets. Building upon this, we propose the Collaborative Static-Dynamic Teaching (CSDT) architecture, a semi-supervised learning architecture that reduces reliance on labeled data by leveraging both static and dynamic teacher models. The framework uses the straight-line prior of stripe-like targets to customize linearity and presents an innovative Adaptive Pseudo-Labeling (APL) strategy, dynamically selecting high-quality pseudo-labels to enhance the student model’s learning process. Extensive experiments on AstroStripeSet and other real-world datasets demonstrate that the CSDT framework achieves state-of-the-art performance in SSTD. Using just 1/16 of the labeled data, CSDT outperforms the second-best Interactive Self-Training Mean Teacher (ISMT) method by 2.64% in mean Intersection over Union (mIoU) and 4.5% in detection rate (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>P</mi><mi>d</mi></msub></semantics></math></inline-formula>), while exhibiting strong generalization in unseen scenarios. This work marks the first application of semi-supervised learning techniques to SSTD, offering a flexible and scalable solution for challenging space imaging tasks.https://www.mdpi.com/2072-4292/17/8/1341stripe-like space target detection (SSTD)semi-supervised learningcollaborative static-dynamic teaching (CSDT)adaptive pseudo-labeling (APL) |
| spellingShingle | Zijian Zhu Ali Zia Xuesong Li Bingbing Dan Yuebo Ma Hongfeng Long Kaili Lu Enhai Liu Rujin Zhao Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection Remote Sensing stripe-like space target detection (SSTD) semi-supervised learning collaborative static-dynamic teaching (CSDT) adaptive pseudo-labeling (APL) |
| title | Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection |
| title_full | Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection |
| title_fullStr | Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection |
| title_full_unstemmed | Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection |
| title_short | Collaborative Static-Dynamic Teaching: A Semi-Supervised Framework for Stripe-like Space Target Detection |
| title_sort | collaborative static dynamic teaching a semi supervised framework for stripe like space target detection |
| topic | stripe-like space target detection (SSTD) semi-supervised learning collaborative static-dynamic teaching (CSDT) adaptive pseudo-labeling (APL) |
| url | https://www.mdpi.com/2072-4292/17/8/1341 |
| work_keys_str_mv | AT zijianzhu collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT alizia collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT xuesongli collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT bingbingdan collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT yueboma collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT hongfenglong collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT kaililu collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT enhailiu collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection AT rujinzhao collaborativestaticdynamicteachingasemisupervisedframeworkforstripelikespacetargetdetection |