激光主动成像图像降噪方法

摘要

According to the characteristics of laser active imaging and the real time requirement,a new image denois-ing method is presented combining lift-wavelet transform and median filter. Firstly,the noise image is dealt with median filter. Secondly, the noise image is decomposed with the lift-wavelet. Finally, the high frequency parts of decomposed image are processed by the threshold function. The image signal to noise ratio ( SNR) and the run time( T) and the image surface chart are applied to estimate the denoising effect of the laser images. The algorithm proposed is compared with the ordinary wavelet, the ordinary wavelet combined with the median filter and so on. Experimental results show that this method can remove the noise of image effectively. Meanwhile,the processing speed is faster. It can meet the requirement of denoise and real-time.%针对激光主动成像图像的特点,并考虑到实时性的要求,提出了一种基于中值滤波与提升小波变换级联的图像降噪方法.首先在空域中对图像进行快速中值滤波,然后对含噪图像进行提升小波分解,最后在小波域内对高频子带进行阈值降噪处理.采用信噪比(SNR)、运行时间T和图像灰度曲面图作为图像降噪效果的评估,将该算法与传统小波降噪、传统小波与中值滤波结合降噪等进行对比实验.实验结果表明,该算法能有效滤除激光图像中的噪声,提高图像的信噪比,且运算速度更快,同时满足了工程应用中对降噪效果和实时性的要求.