阈值优化的遥感影像小波去噪

Threshold Optimized Wavelet for Remotely Sensed Image Denoising

针对传统的小波阈值在去除遥感影像噪声时存在噪声残留和噪声误判的问题,提出了针对遥感影像的小波阈值函数优化算法。该算法利用小波边缘检测算法确定遥感影像边缘特征的小波系数,然后根据噪声的方差设置优化的阈值函数去噪,即在以往的统一阈值基础上加以修改,使阈值能随着分解尺度的变化而改变,对传统的软阈值和硬阈值的优点予以保留,改进它们的缺点,生成一种新的阈值函数,使它在处理小波系数时更加灵活。经过优化的小波阈值去噪后得到平滑遥感影像,之后把小波边缘检测遥感影像镶嵌入平滑遥感影像中。实验表明,与传统的小波阈值去噪方法相比,该算法解决了传统阈值函数在去噪过程中出现的问题,进行有效去噪的同时保留遥感影像的细节,使影像更加清晰,提高了信噪比。

To solve the problems of denoising limitation, noise residue and noise misjudgment of traditional wavelet threshold in removing noise of remote sensing image, an optimization algorithm of wavelet threshold function in view of the remote sensing image was brought forward. The algorithm made use of wavelet edge detection algorithm to determine the wavelet coefficients of remote sensing image edge character. Then, according to the noise variance, it set optimized threshold function to remove noise,in other words,it modified the noise variance and made it vary with the decomposition scale based on the previous unified threshold. The algorithm preserved advantages of the traditional soft threshold and hard threshold, improved their defects,and generated a new threshold function,while making the new threshold more flexible in the processing of wavelet coefficients. After optimization of wavelet threshold denoising, the smooth remote sensing image could be obtained. Finally, the remote sensing image of wavelet edge detection was embedded into the smooth remote sensing image. The experimental results show that compared with the traditional wavelet threshold denoising method,the algorithm could solve the denoising problems in the traditional threshold function,keep the details of the remote sensing image in the course of removing noise, and improve the signal-to-noise ratio.