摘要
提出了一种基于小波变换理论的超分辨率重建算法,即利用小波变换得到图像的高频和低频子带,结合非线性外推技术对高频子带进行处理,在增加高频子带信息量的同时进行迭代改进,并采用小波阈值方法进行去噪处理。实验结果表明:该算法能够克服以往插值算法的不足,如高频损失、细节模糊等,能很好地提高图像的峰值信噪比,是图像重建的一种有效方法。
Abstr act： Super-resolution image reconstruction refers to a resolution enhancement technology,which reconstruct highresolution image from the related low-resolution. This technique does not need to change the existing imaging equipment, which has important research implications. In this paper,an image Super-resolution reconstruction algorithm based on wavelet transform is provided. Wavelet transform can separate high frequency and low frequency information. By combining nonlinear extrapolation the more high frequency information is gained. Meanwhile, the iterafive method is used to reconstruct the image and the method of wavelet threshold de-noising is used. The experiment results show that this algorithm overcome the disadvantage of the classical interpolation method, such as high frequency image loss and details blurry, which improves the resolution and PSNR. It is a useful method of image super-resolution reconstruction. Key words ： super-resolution ; wavelet transform ; iteration ; nonlinear extrapolation ; threshold de-noising
出处
《天津职业技术师范大学学报》
2014年第1期45-49,共5页
Journal of Tianjin University of Technology and Education
基金
天津市应用基础及前沿技术研究计划项目(12JCYBJC10600)
天津职业技术师范大学研究生创新基金资助项目
关键词
超分辨率重建
小波变换
迭代
非线性外推
阈值去噪
super-resolution
wavelet transform
iteration
nonlinear extrapolation
threshold de-noising
