国产高分辨率遥感影像辐射质量改善方法研究

Study on Improvement of Domestic High Resolution Satellite Image Radiation Quality

受制于光学器件的性能与观测环境的影响,国产高分辨率遥感影像通常存在条带噪声、光谱分辨率低、纹理不清晰等问题,给后续的自动化影像产品生产及目标解译工作带来极大的不利。本研究先后从去除条带噪声、对比度增强、边缘增强三个方面,对遥感影像进行质量改善,使遥感影像更好地为地理信息公众服务。以高分二号全色影像为实验数据,目视和定量评价结果均证实了本文算法的有效性。

Limited by performance of optics and observation environment,domestic high resolution satellite images are bothered by strip noise,low spectral resolution and fuzzy texture,bringing adverse effect to the following automated image products production and objective interpretation. This study improves the quality of remote sensing images from three aspects,strip noise removal,contrast enhancement and edge enhancement,to make remote sensing images better serve the public platform of geographic information. Using the GF-2 panchromatic image as test data,the visual results and the quantitative evaluation results verified the effectiveness of the proposed method.