摘要
讨论一种新的图像分辨率增强的偏微分方程模型。新模型利用图像为分片连续的二维曲面的先验信息,采用非线性扩散方程对基于最近邻方法放大后的图像进行光滑以消除因放大引起的“块状”效应,同时在迭代过程中加入插值条件使处理后的图像能较好的保留边缘,由此构造出基于偏微分方程的图像分辨率增强模型和相应的迭代算法,使图像的分辨率增强到亚像元级。遥感图像的仿真计算结果表明,该方法具有比现有方法更高的峰值信噪比和更高的边缘保护指数。
This paper discusses a new partial differential equation model for resolution enhancement of image. This model adapts the prior information of piecewise continuity of image to constiuct a nonlinear diffusion model, which is used to smooth an enlarged image by nearest interpolation to reduce the blocky effect caused by enlargement, simultaneity, and interpolation condition is added into the iterated process to avoid a too smooth image. The algorithm is tested by three remote sensing images, all of the results show that the new model has higher peak signal to noise ratio and higher edge preserve index than the known methods.
出处
《遥感学报》
EI
CSCD
北大核心
2005年第6期673-679,共7页
NATIONAL REMOTE SENSING BULLETIN
基金
国家自然科学基金(项目号:60272013)
全国优秀博士论文作者专项基金(项目号:200140)
关键词
分辨率增强
插值
偏微分方程
超分辨
遥感
resolution enhancement
interpolation
partial differential equation
remote sensing
