遥感图像质量提升研究

发布时间：2018-02-04 17:56

本文关键词： 图像去噪条带噪声图像重建相位一致性刃边法 PSF FTVD　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：卫星遥感成像过程中，大气、光学系统以及传感器等多种因素综合作用，导致图像退化严重，解析度降低，进而影响卫星遥感图像的使用。本文以提升遥感图像质量为目的，建立了遥感图像退化模型，从噪声去除和图像复原两方面入手，设计了有效的提升遥感图像质量的重建算法。首先，，针对影响遥感图像最为严重的条带噪声展开深入分析，设计了改进的矩匹配法，对遥感图像中宽条带的非均匀性进行校正，再利用基于单向变分模型优化算法对条带噪声进行去除。为提高遥感图像的解析度，设计了基于相位一致性与Hough变换技术相结合的改进刃边法，实现了对遥感图像的点扩散函数(PSF)的精确估计，利用快速全变分去卷积法(FTVD)对遥感图像进行重建，提升了遥感图像的质量。通过对处理后的遥感图像进行像质评价分析，结果表明本文提出的算法可以有效地提高卫星遥感图像的分辨率和信噪比，改善了图像的质量。论文还分析了不同因素对重建结果的影响，为进一步提升遥感图像质量的研究提供了参考。
[Abstract]:In the process of satellite remote sensing imaging, many factors, such as atmosphere, optical system and sensor, make the image degenerate seriously and the resolution decreases. In order to improve the quality of remote sensing image, the degradation model of remote sensing image is established, which starts from two aspects: noise removal and image restoration. An effective reconstruction algorithm is designed to improve the quality of remote sensing image. Firstly, the improved moment matching method is designed to correct the non-uniformity of wide strip in remote sensing image. In order to improve the resolution of remote sensing image, an improved edge method based on phase consistency and Hough transform is designed. The point diffusion function (PSF) of remote sensing image is estimated accurately, and the fast total variational deconvolution method is used to reconstruct the remote sensing image. Through the image quality evaluation and analysis of processed remote sensing images, the results show that the proposed algorithm can effectively improve the resolution and signal-to-noise ratio of satellite remote sensing images. The paper also analyzes the influence of different factors on the reconstruction results, which provides a reference for further improving the quality of remote sensing images.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP751

【参考文献】