【摘要】 给出了一种构造双正交小波滤波器的设计方法,应用该方法得到的双正交小波对低分辨率图像进行小波分解。对分解后的图像结构进行分析,提出双三次插值与方向插值相结合的方法,对分解后图像插值。对插值后图像小波逆变换,得到超分辨率图像。应用于CCD图像,获得了比双线性插值高的峰值信噪比(25.3090dB),图像细节信息增强了。实验结果表明,算法应用于CCD图像不仅提高了图像的空间分辨率,同时也提高了图像的峰值信噪比,还比较好地保留了图像的边缘信息,且经过该方法恢复后的图像更适合人眼观察,细节丰富,更加清晰,畸变小,是一种提高CCD图像空间分辨率的有效方法。更多还原

【Abstract】 A design method of bi-orthogonal wavelet filter is proposed.Biorthogonal wavelet produced is applied to decompose the low-resolution image.By analyzing the structure of the decomposed image,the combination method of bi-cubic interpolation and the direction interpolation is proposed.This method is used to interpolate the decomposed image.The super-resolution image is got after transforming the interpolated image by the inverse wavelet.Applied the method to CCD images,higher PSNR(25.3090dB) than bilinear interpolation are obtained.The image detail information is enhanced.The results show that the algorithm is applied to CCD images not only can improve the image’s spatial resolution,but also can improve the peak signal to noise ratio of the image.It is better to retain the image edge information.The image restored through this method is more suitable for the human eye observation,the details are richer and clearer,distortion is small.The method shows that the method is an effective way in improving spatial resolution of CCD image.更多还原