倒易晶胞和小波相结合的图像复原方法  

Image Restoration Based on Hybrid Reciprocal Cell-Wavelet

作　　者：张智[1] 夏德深[1] 

机构地区：[1]南京理工大学计算机科学与技术学院,南京210094

出　　处：《计算机辅助设计与图形学学报》2008年第4期512-519,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：香港特区政府研究资助局项目(CUHK/4180/01E);江苏省教育厅自然科学基金(04KJD520037)

摘　　要：为了去除混迭、噪声和模糊对图像的影响,从遥感器成像的光学及几何特性出发,运用倒易晶胞的复原方法对晶胞形状进行限制;将错误的频谱覆盖放到正确的位置以减少混迭;并结合小波复原方法对退化的图像进行去模糊、去噪,得到一幅更清晰的图像.实验结果表明,结合倒易晶胞的小波复原方法比常用基于频域的方法复原效果更好.Based on the optic and geometrical characteristics of the remote sensing image acquisition system, a reciprocal cell restoration is proposed for image antialiasing in this work. As a result, the shape of the unit cell could be confined, and then the superposition of the frequency could be removed. Thereafter, the wavelet restoration is applied to deblur and denoise the image. The results of the test indicate that the proposed restoration approach is better than the conventional methods of frequency restorations.

关 键 词：混迭 复原 小波 倒易晶胞 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]