Scholarlink

Aiming at the strip noise in the overlap region of a multi-spectral image of the space optical remote sensing camera with reflective mirror splicing, it is proposed in this paper that a normalized synchronization method of distortion correction and radiometric information of homonymic pixels in the overlap region. Pixels in the overlap region. To solve the problem of strip noise in the overlap region of the multi-spectral image, the radiometric models are both reconstructed by adding the radiometric models in the overlap region. Reconstructed by adding the radiometric energy of homonymic pixels after the distortion corrections of homonymic pixels in the overlap region, and the moment matching method is used to treat the edge of the overlap region with transitional uniform color. At the same time, the fringe coefficient is optimized, the fringe coefficient is optimized so that the fringe coefficient is more sensitive to the fringe region and has no effect on the non-fringe region, and the fringe energy analysis is quantified. Experimental results show that the strip noise in the overlap is reduced or even disappears, the fringe coefficient declines, and the radiometric uniformity of the multi-spectrum is reduced. Experimental results show that the strip noise in the overlap is reduced or even disappeared, fringe coefficient declined and the radiometric uniformity of multi-spectral images is enhanced significantly.

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

An Optical Remote Sensing Satellite Relative Radiometric Calibration Method Based on Statistical Sample Weighting

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Huaying He,
  • Chao Deng,
  • Xiaoyu Huang,
  • Ao Zhang,
  • Jian Zeng,
  • Qijin Han,
  • Yu Wu

摘要

Aiming at the strip noise in the overlap region of a multi-spectral image of the space optical remote sensing camera with reflective mirror splicing, it is proposed in this paper that a normalized synchronization method of distortion correction and radiometric information of homonymic pixels in the overlap region. Pixels in the overlap region. To solve the problem of strip noise in the overlap region of the multi-spectral image, the radiometric models are both reconstructed by adding the radiometric models in the overlap region. Reconstructed by adding the radiometric energy of homonymic pixels after the distortion corrections of homonymic pixels in the overlap region, and the moment matching method is used to treat the edge of the overlap region with transitional uniform color. At the same time, the fringe coefficient is optimized, the fringe coefficient is optimized so that the fringe coefficient is more sensitive to the fringe region and has no effect on the non-fringe region, and the fringe energy analysis is quantified. Experimental results show that the strip noise in the overlap is reduced or even disappears, the fringe coefficient declines, and the radiometric uniformity of the multi-spectrum is reduced. Experimental results show that the strip noise in the overlap is reduced or even disappeared, fringe coefficient declined and the radiometric uniformity of multi-spectral images is enhanced significantly.