小波超分辨率重建算法及其在SPOT影像中的应用

影像超分辨率重建就是从一系列质量较差、分辨率较低的图像来重建图像质量更好、空间分辨率更高的影像的算法.在将所有低分辨率影像纳入同一参考格网中考虑的情况下,低分辨率数据相当于一非规则采样的交错采样数据.本文在多分辨率分析的基础上对基于小波的影像超分辨率重建理论进行介绍,并将现有的小波超分辨率重建算法推广到更一般的运动模型,以充分利用交错采样数据的内在规律性和结构特征,同时针对SPOT影像的特点,分析SPOT影像的调制传送函数,对SPOT重建影像进行包括去噪与解卷积的复原后处理.最后给出基于小波的SPOT影像超分辨率重建结果,表明遥感影像的超分辨率重建在应用中取得进展.

Wavelet Super-Resolution Algorithm and Its Application to SPOT Images

Image superresolution refers to image processing algorithms which produce the images of high quality and high-resolution (HR) from a set of images of low quality and low-resolution (LR). When considering all these given LR images on a reference grid, they are nonuniformly interlaced sampled. Based on multiresolution analysis, the wavelet superresolution algorithm is introduced and extended to more general camera motion model, which can take advantage of the regularity and structure inherent in interlaced data. For the special case of SPOT images superresolution, the MTF of SPOT image is analyzed and the post-processing including denoising and deconvolution. for reconstructed SPOT image is proposed.Finally,some examples of wavelet superresolution for SPOT image are given. It announces the progress made in the application of superresolution algorithm for remote sensing images.