一种基于小波与双三次插值的CCD图像超分辨方法  

Kind of super-resolution method of CCD image based on wavelet and bicubic interpolation

作　　者：赵秀影[1,2] 苏耘[2] 董艳芹[2] 王敬梅 翟林培[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所,长春130033 [2]空军航空大学,长春130022 [3]青岛城市建设投资(集团)有限责任公司,山东青岛266071

出　　处：《计算机应用研究》2009年第6期2365-2367,共3页Application Research of Computers

基　　金：国家“863”计划资助项目(2007AA701203)

摘　　要：为了尽可能地保持CCD图像的原始信息,提高图像的空间分辨率,有利于对图像的细节信息进行观察分析,对各种超分辨方法进行研究,提出了一种改进的基于小波和双三次插值的超分辨方法:对低分辨率图像进行灰度变换,并把它作为小波逆变换的低分辨率图像,对图像进行恢复,再与低分辨图像的双三次插值图像求平均。将该方法应用于CCD图像,从视觉上空间分辨率有提高,并可以获得25.524 4 dB的峰值信噪比。实验结果表明,该算法得到了比全小波双三次插值、原图像作为低频图像小波双三次插值和双线性插值更高的峰值信噪比及更好的图像细节效果,是一种适合提高CCD图像分辨率的有效方法。In order to maintain the original information of the CCD images, to improve the spatial resolution of image, to observe and analyze the details of the images, this paper studied on various super-resolution methods. This paper proposed the improved super-resolution method based on wavelets and bicubic interpolation. It transformed low-resolution images of grayscale, and considered it as the low-resolution image in an inverse wavelet transform, recoveried the super-resolution image, then got an average of with the image of low-resolution ＇ s bicubic interpolation. Applied this method in CCD images, improved the spatial resolution in the visual, and made the peak signal to noise ratio 25. 524 4 dB. Experimental results show that using this method in low-resolution image can get higher PSNR and better image＇ s details effect than those only using bicubic interpolation in the whole wavelet domain or those using the original image as the low-frequency image in an inverse wavelet transform of the bicubic interpolation or those only using the bilinear. The experiment results show that this method is an effective way for improving the CCD image＇s resolution.

关 键 词：CCD图像 超分辨 小波变换 双三次插值 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]