Scholarlink

<p>Block adjustment is crucial for large-scale elevation calibration of interferometric synthetic aperture radar digital surface models (DSMs). General polynomial model struggles to accurately depict the propagation law of interferometric parameter error. To this end, an interferometric parameter block adjustment method was proposed. Sensitivity equations were expanded to construct the parametric adjustment model, which was solved using the enriched Krylov subspace method. Ice, Cloud, and Land Elevation Satellite (ICESat)-2 points served as control and checkpoints, with Copernicus DSM30 for qualitative comparison. Experiments with TanDEM-X bistatic data near Guangzhou, China, indicate that (1) the bias introduced from sensitivity equation expansion was approximately 1e−6. (2) The least-square solution underestimated elevations in mountainous regions (root-mean-square error (RMSE): 3.64&#xa0;m), and the conjugate gradient solution retained systematic errors in certain DSMs (RMSE: 3.58&#xa0;m). Contrarily, the enriched Krylov subspace solution yielded more accurate and reasonable results owing to its consideration of error characteristic. (3) Compared to the polynomial model, the parametric model improved the elevation of each DSM, particularly in steep areas, reducing mean, RMSE, and 90% linear error against ICESat-2 by 1.35&#xa0;m, 0.44&#xa0;m, and 1.16&#xa0;m, respectively, with final values of 0.25&#xa0;m, 3.49&#xa0;m, and 5.01&#xa0;m. Key profile lines aligned closer to the benchmark DSM. These results demonstrate the applicability of the proposed method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enriched Krylov subspace method for resolving interferometric parameter block adjustment model in elevation calibration of spaceborne InSAR digital surface model

  • 期刊论文

  • 发表日期：

    2025年7月23日

查看全文
  • Liqun Liu,
  • Haiqiang Fu,
  • Chenglong Cao,
  • Yifan Zhang,
  • Kun Han,
  • Xun Du,
  • Pengfei Li,
  • Jie Li,
  • Zhiwei Li

摘要

Block adjustment is crucial for large-scale elevation calibration of interferometric synthetic aperture radar digital surface models (DSMs). General polynomial model struggles to accurately depict the propagation law of interferometric parameter error. To this end, an interferometric parameter block adjustment method was proposed. Sensitivity equations were expanded to construct the parametric adjustment model, which was solved using the enriched Krylov subspace method. Ice, Cloud, and Land Elevation Satellite (ICESat)-2 points served as control and checkpoints, with Copernicus DSM30 for qualitative comparison. Experiments with TanDEM-X bistatic data near Guangzhou, China, indicate that (1) the bias introduced from sensitivity equation expansion was approximately 1e−6. (2) The least-square solution underestimated elevations in mountainous regions (root-mean-square error (RMSE): 3.64 m), and the conjugate gradient solution retained systematic errors in certain DSMs (RMSE: 3.58 m). Contrarily, the enriched Krylov subspace solution yielded more accurate and reasonable results owing to its consideration of error characteristic. (3) Compared to the polynomial model, the parametric model improved the elevation of each DSM, particularly in steep areas, reducing mean, RMSE, and 90% linear error against ICESat-2 by 1.35 m, 0.44 m, and 1.16 m, respectively, with final values of 0.25 m, 3.49 m, and 5.01 m. Key profile lines aligned closer to the benchmark DSM. These results demonstrate the applicability of the proposed method.