High-Precision Reconstruction of Water Areas Based on High-Resolution Stereo Pairs of Satellite Images
Abstract
1. Introduction
- This article proposes a water surface stereo reconstruction algorithm named boundary plane-constrained surface water stereo reconstruction (BPC-SWSR), which uses effective shoreline matching points to optimize the reconstruction of the weak texture regions, eliminating serious issues such as significant holes and abnormal undulations, while achieving smoother and more accurate water surface reconstructions.
- The algorithm proposed in this article is based on plane and tilt angle constraints, while employing multiple strategies to achieve the seamless planar reconstruction of water bodies of various types and sizes, demonstrating strong versatility in the field of water area reconstruction.
- We conduct experiments to compare the accuracy and smoothness of our algorithm reconstruction results with the results of the current state-of-the-art reconstruction algorithms. Our method achieves a significantly lower RMSE of 2.279 m versus 302.4 m for the other methods. Moreover, our method achieves a variance of 0.6613 m2 compared with the average variance of 522.5 m2 for the other methods. DSMs generated by our algorithm are much more accurate and smoother in all kinds of water regions.
2. Related Work
2.1. Stereo Dense Matching
2.2. Stereo Dense Matching Algorithms for Weak Texture Areas
2.2.1. Adaptive Selection of Matching Cost Windows
2.2.2. Matching Propagation Enhancement Method
2.2.3. Geometric Relationship Constraint Method
3. Method
- Effective shoreline area extraction: this article uses the DeepLabV3+ [18] pretrained model to extract water surfaces, buildings, and vegetation masks from multispectral satellite imagery corresponding to the target locations. After morphological processing, an initial shoreline area is obtained. Multiple strategies are employed to derive an effective shoreline ground mask.
- Base water surface fitting: On the basis of the effective shoreline data, a RANSAC [35] plane fitting strategy with slope and inclination constraints is applied to assign values to the water surface region according to the fitted plane parameters, resulting in a base water surface. The original disparity image is generated by SGM [22]. The time complexity of this part is , where n denotes the number of pixels in the disparity image of the target water region. This complexity arises because, after the plane is fitted using a small set of control points (whose number is negligible relative to n), each pixel within the target water region is assigned a disparity value based on a simple planar equation. This per-pixel assignment requires constant time, leading to a total linear complexity with respect to the number of pixels.
- Seamless water surface reconstruction: Starting with the base water surface as the initial value, a second-order smoothing water surface reconstruction model is established to obtain a seamlessly connected disparity-optimized result. A high-precision water surface DSM is generated on the basis of the optimized disparity image. This part achieves linear time complexity , with n representing the pixel count in the water region’s disparity image. The linear complexity arises because the reconstruction model operates directly on each pixel in the disparity map using local neighborhood information (e.g., second-order derivatives or convolutional filters). These operations involve a fixed number of computations per pixel, resulting in a total processing time that scales linearly with the number of pixels in the target region.
3.1. Shoreline Area Extraction
3.1.1. Water, Vegetation, and Building Mask Extraction
3.1.2. Effective Shoreline Area Extraction
3.2. Base Water Surface Fitting
3.3. Seamless Water Surface Reconstruction
4. Experiments
4.1. Datasets
4.1.1. Beijing-3 Images
4.1.2. GF-7 Images
4.1.3. Pleiades Images
4.1.4. SuperView Images
4.1.5. WV-3 Images
4.2. Implementation and Evaluation Metrics
4.2.1. Implementation Details
4.2.2. Evaluation Metrics
4.3. Comparison with Other Methods
4.3.1. Quantitative Comparison
4.3.2. Qualitative Comparison
5. Discussion
5.1. Effect of Removing Shoreline Vegetation and Buildings Areas
5.2. Effect of Seamless Water Surface Reconstruction
5.3. Influence of the Global Plane Fitting Strategy
5.4. Influence of the Virtual Shoreline Strategy
5.5. Advantages and Limitation Analysis
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
BPC-SWSR | boundary plane-constrained surface water stereo reconstruction |
RM | Reconstruction Master version 6.0 |
RMSE | Root mean squared error |
ME | Mean error |
VAR | Variance |
DSM | Digital surface model |
References
- El Garouani, A.; Alobeid, A.; El Garouani, S. Digital surface model based on aerial image stereo pairs for 3D building. Int. J. Sustain. Built Environ. 2014, 3, 119–126. [Google Scholar] [CrossRef]
- Guerin, C.; Binet, R.; Pierrot-Deseilligny, M. Automatic Detection of Elevation Changes by Differential DSM Analysis: Application to Urban Areas. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2014, 7, 4020–4037. [Google Scholar] [CrossRef]
- Zhou, Q. Digital elevation model and digital surface model. Int. Encycl. Geogr. People Earth Environ. Technol. 2017, 1–17. [Google Scholar] [CrossRef]
- Yuan, X.; Tian, J.; Reinartz, P. A Self-Training Approach Using Benchmark Dataset and Stereo-DSM for Building Extraction. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2024, 17, 11352–11364. [Google Scholar] [CrossRef]
- Brunn, A.; Weidner, U. Extracting buildings from digital surface models. Int. Arch. Photogramm. Remote Sens. 1997, 32, 27–34. [Google Scholar]
- Altmaier, A.; Kany, C. Digital surface model generation from CORONA satellite images. ISPRS J. Photogramm. Remote Sens. 2002, 56, 221–235. [Google Scholar] [CrossRef]
- White, J.C.; Wulder, M.A.; Vastaranta, M.; Coops, N.C.; Pitt, D.; Woods, M. The utility of image-based point clouds for forest inventory: A comparison with airborne laser scanning. Forests 2013, 4, 518–536. [Google Scholar] [CrossRef]
- Zhang, L. Automatic Digital Surface Model (DSM) Generation from Linear Array Images; ETH: Zurich, Switzerland, 2005. [Google Scholar]
- Zheng, Y.; Yang, S.; Li, Y.; Wu, J.; Shi, Z.; Kou, R. A Heterogeneous Remote Sensing Image Matching Method for Urban Areas with Complex Terrain Based on 3D Spatial Relationship Constraints. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2024, 17, 6791–6804. [Google Scholar] [CrossRef]
- Adhyapak, S.A.; Kehtarnavaz, N.; Nadin, M. Stereo matching via selective multiple windows. J. Electron. Imaging 2007, 16, 013012. [Google Scholar] [CrossRef]
- Huang, X.; Zhang, Y.; Yue, Z. Image-guided non-local dense matching with three-steps optimization. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 2016, 3, 67–74. [Google Scholar]
- Wu, B.; Zhang, Y.; Zhu, Q. A triangulation-based hierarchical image matching method for wide-baseline images. Photogramm. Eng. Remote Sens. 2011, 77, 695–708. [Google Scholar] [CrossRef]
- Xu, G.; Wang, X.; Zhang, Z.; Cheng, J.; Liao, C.; Yang, X. IGEV++: Iterative Multi-range Geometry Encoding Volumes for Stereo Matching. IEEE Trans. Pattern Anal. Mach. Intell. 2025, 1–15. [Google Scholar] [CrossRef]
- Xu, G.; Cheng, J.; Guo, P.; Yang, X. Attention concatenation volume for accurate and efficient stereo matching. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA, 18–24 June 2022; pp. 12981–12990. [Google Scholar]
- Chen, K.; Chen, B.; Liu, C.; Li, W.; Zou, Z.; Shi, Z. Rsmamba: Remote sensing image classification with state space model. IEEE Geosci. Remote Sens. Lett. 2024, 21, 8002605. [Google Scholar] [CrossRef]
- Zheng, S.; Lu, C.; Wu, Y.; Gupta, G. SAPNet: Segmentation-aware progressive network for perceptual contrastive deraining. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, HI, USA, 3–8 January 2022; pp. 52–62. [Google Scholar]
- Li, X.; Xu, F.; Liu, F.; Tong, Y.; Lyu, X.; Zhou, J. Semantic segmentation of remote sensing images by interactive representation refinement and geometric prior-guided inference. IEEE Trans. Geosci. Remote Sens. 2023, 62, 5400318. [Google Scholar] [CrossRef]
- Chen, L.C.; Zhu, Y.; Papandreou, G.; Schroff, F.; Adam, H. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, 8–14 September 2018; pp. 801–818. [Google Scholar]
- Scharstein, D.; Szeliski, R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. Int. J. Comput. Vis. 2002, 47, 7–42. [Google Scholar] [CrossRef]
- Sha, H.; Yuan, X. The Review and Prospect of Dense Matching Methods for Stereo Imagery. Geomat. Inf. Sci. Wuhan Univ. 2023, 48, 1813–1833. [Google Scholar]
- Zhu, C.T.; Chang, Y.Z.; Wang, H.M.; He, K.; Lee, S.T.; Lee, C.F. Efficient stereo matching with decoupled dissimilarity measure using successive weighted summation. Math. Probl. Eng. 2014, 2014, 127284. [Google Scholar] [CrossRef]
- Huang, X.; Han, Y.; Hu, K. An improved semi-global matching method with optimized matching aggregation constraint. In IOP Conference Series: Earth and Environmental Science; IOP Publishing: Bristol, UK, 2020; Volume 569, p. 012050. [Google Scholar]
- Xiang, J.; Li, Z.; Kim, H.S.; Chakrabarti, C. Hardware-efficient neighbor-guided SGM optical flow for low power vision applications. In Proceedings of the 2016 IEEE International Workshop on Signal Processing Systems (SiPS), Dallas, TX, USA, 26–28 October 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–6. [Google Scholar]
- Hirschmüller, H.; Buder, M.; Ernst, I. Memory efficient semi-global matching. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 2012, 1, 371–376. [Google Scholar] [CrossRef]
- Kanade, T.; Okutomi, M. A stereo matching algorithm with an adaptive window: Theory and experiment. IEEE Trans. Pattern Anal. Mach. Intell. 1994, 16, 920–932. [Google Scholar] [CrossRef]
- Zhang, K.; Lu, J.; Lafruit, G. Cross-based local stereo matching using orthogonal integral images. IEEE Trans. Circuits Syst. Video Technol. 2009, 19, 1073–1079. [Google Scholar] [CrossRef]
- Emlek, A.; Peker, M.; Dilaver, K.F. Variable window size for stereo image matching based on edge information. In Proceedings of the 2017 International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, Turkey, 16–17 September 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 1–4. [Google Scholar]
- Huang, X.; Liu, W.; Qin, R. A window size selection network for stereo dense image matching. Int. J. Remote Sens. 2020, 41, 4838–4848. [Google Scholar] [CrossRef]
- Gan, Y.; Hamzah, R.; Anwar, N.N. Local stereo matching algorithm based on pixel difference adjustment, minimum spanning tree and weighted median filter. In Proceedings of the 2018 IEEE Conference on Systems, Process and Control (ICSPC), Melaka, Malaysia, 14–15 December 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 39–43. [Google Scholar]
- Mozerov, M.G.; Van De Weijer, J. Accurate stereo matching by two-step energy minimization. IEEE Trans. Image Process. 2015, 24, 1153–1163. [Google Scholar] [CrossRef] [PubMed]
- Zhang, Y. A Multi-Primitive Multi-View Image Matching Method with Adaptive Triangular Constraints. Ph.D. Thesis, Wuhan University, Wuhan, China, 2011. [Google Scholar]
- Wu, B.; Zhang, Y.; Zhu, Q. Integrated point and edge matching on poor textural images constrained by self-adaptive triangulations. ISPRS J. Photogramm. Remote Sens. 2012, 68, 40–55. [Google Scholar] [CrossRef]
- Deng, C.; Liu, D.; Zhang, H.; Li, J.; Shi, B. Semi-global stereo matching algorithm based on multi-scale information fusion. Appl. Sci. 2023, 13, 1027. [Google Scholar] [CrossRef]
- Jia, L.; Wang, C.; Wang, H. Comparative Analysis of 3D Modeling Software for Drone Oblique Photography. Geomat. Spat. Inf. Technol. 2024, 47, 176–179. [Google Scholar]
- Yang, L.; Li, Y.; Li, X.; Meng, Z.; Luo, H. Efficient plane extraction using normal estimation and RANSAC from 3D point cloud. Comput. Stand. Interfaces 2022, 82, 103608. [Google Scholar] [CrossRef]
- Wang, J.; Zheng, Z.; Ma, A.; Lu, X.; Zhong, Y. LoveDA: A remote sensing land-cover dataset for domain adaptive semantic segmentation. arXiv 2021, arXiv:2110.08733. [Google Scholar]
- Morel, J.M.; Petro, A.B.; Sbert, C. Fourier implementation of Poisson image editing. Pattern Recognit. Lett. 2012, 33, 342–348. [Google Scholar] [CrossRef]
- Zhang, J.; Liu, Y.; Chen, L. Research and Application of Large-Scale Mapping Technology Based on Beijing-3 Stereo Imagery. Bull. Surv. Mapp. 2024, 49–52. [Google Scholar]
- Zhou, P.; Tang, X. Geometric Accuracy Verification of GF-7 Satellite Stereo Imagery without GCPs. IEEE Geosci. Remote Sens. Lett. 2022, 19, 6509105. [Google Scholar] [CrossRef]
- Boissin, M.B.; Gleyzes, A.; Tinel, C. The pléiades system and data distribution. In Proceedings of the 2012 IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany, 22–27 July 2012; pp. 7098–7101. [Google Scholar] [CrossRef]
- Liu, Y.K.; Ma, L.L.; Wang, N.; Qian, Y.G.; Zhao, Y.G.; Qiu, S.; Gao, C.X.; Long, X.X.; Li, C.R. On-orbit radiometric calibration of the optical sensors on-board SuperView-1 satellite using three independent methods. Opt. Express 2020, 28, 11085–11105. [Google Scholar] [CrossRef] [PubMed]
- Tong, F.; Tong, H.; Mishra, R.; Zhang, Y. Delineation of Individual Tree Crowns Using High Spatial Resolution Multispectral WorldView-3 Satellite Imagery. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2021, 14, 7751–7761. [Google Scholar] [CrossRef]
- Guo, X.; Yang, K.; Yang, W.; Wang, X.; Li, H. Group-wise correlation stereo network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 15–20 June 2019; pp. 3273–3282. [Google Scholar]
- Fei, B.; Yang, W.; Chen, W.; Ma, L.; Hu, X. Hff-net: Hierarchical feature fusion network for point cloud generation with point transformers. In Proceedings of the 2022 IEEE International Conference on Multimedia and Expo (ICME), Taipei, Taiwan, 18–20 July 2022; IEEE: Piscataway, NJ, USA, 2022; pp. 1–6. [Google Scholar]
- Chang, J.R.; Chen, Y.S. Pyramid stereo matching network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18–23 June 2018; pp. 5410–5418. [Google Scholar]
- Wei, K.; Huang, X.; Li, H. Stereo Matching Method for Remote Sensing Images Based on Attention and Scale Fusion. Remote Sens. 2024, 16, 387. [Google Scholar] [CrossRef]
Methods | RIVER | LAKE | SEA | AVERAGE | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
RMSE | ME | VAR | RMSE | ME | VAR | RMSE | ME | VAR | RMSE | ME | VAR | |
ACVNet [14] | 27.08 | 20.40 | 651.6 | 50.31 | 43.55 | 856.5 | 1279 | 1273 | 771.1 | 452.3 | 445.6 | 759.7 |
GWC [43] | 37.22 | 28.49 | 997.6 | 61.55 | 45.64 | 1650 | 1274 | 1263 | 2491 | 457.6 | 445.8 | 1713 |
HFF [44] | 18.80 | 13.29 | 411.4 | 15.65 | 9.471 | 261.2 | 1284 | 1275 | 609.5 | 439.6 | 432.9 | 427.4 |
PSM [45] | 11.52 | 7.698 | 176.8 | 12.80 | 7.914 | 166.6 | 1272 | 1270 | 219.1 | 432.2 | 428.7 | 187.6 |
ASF [46] | 8.883 | 7.517 | 44.09 | 15.62 | 14.43 | 45.50 | 13.19 | 10.41 | 147.2 | 11.76 | 10.17 | 65.14 |
IGEV++ [13] | 6.537 | 5.405 | 25.96 | 10.42 | 9.742 | 21.01 | 6.014 | 5.470 | 6.989 | 7.596 | 6.719 | 20.68 |
SGM [22] | 14.68 | 8.733 | 311.5 | 17.75 | 9.959 | 364.2 | 42.02 | 25.72 | 2226 | 24.82 | 14.81 | 967.1 |
RM [34] | 4.418 | 2.279 | 19.14 | 13.96 | 11.32 | 77.97 | 5.054 | 3.322 | 20.42 | 7.812 | 5.643 | 39.18 |
Ours | 2.209 | 2.089 | 0.662 | 3.283 | 2.841 | 1.221 | 1.345 | 0.984 | 0.101 | 2.279 | 1.971 | 0.661 |
WF 1 | BF 2 | PB 3 | RMSE (m) | ME (m) | VAR (m2) |
---|---|---|---|---|---|
24.82 | 14.81 | 967.1 | |||
✓ | 7.081 | 5.142 | 2.547 | ||
✓ | ✓ | 2.330 | 2.045 | 0.675 | |
✓ | ✓ | ✓ | 2.279 | 1.971 | 0.661 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ye, J.; Xu, R.; Wang, Y.; Huang, X. High-Precision Reconstruction of Water Areas Based on High-Resolution Stereo Pairs of Satellite Images. Remote Sens. 2025, 17, 2139. https://doi.org/10.3390/rs17132139
Ye J, Xu R, Wang Y, Huang X. High-Precision Reconstruction of Water Areas Based on High-Resolution Stereo Pairs of Satellite Images. Remote Sensing. 2025; 17(13):2139. https://doi.org/10.3390/rs17132139
Chicago/Turabian StyleYe, Junyan, Ruiqiu Xu, Yixiao Wang, and Xu Huang. 2025. "High-Precision Reconstruction of Water Areas Based on High-Resolution Stereo Pairs of Satellite Images" Remote Sensing 17, no. 13: 2139. https://doi.org/10.3390/rs17132139
APA StyleYe, J., Xu, R., Wang, Y., & Huang, X. (2025). High-Precision Reconstruction of Water Areas Based on High-Resolution Stereo Pairs of Satellite Images. Remote Sensing, 17(13), 2139. https://doi.org/10.3390/rs17132139