噪声模糊图像复原算法构建及仿真  

Construction and Simulation of Noise Blurred Image Restoration Algorithm

作　　者：孙文静[1] SUN Wen-jing(Dianchi College Yunnan University,Kunming 650228 China)

机构地区：[1]云南大学滇池学院,云南昆明650228

出　　处：《自动化技术与应用》2023年第11期30-33,共4页Techniques of Automation and Applications

基　　金：云南省教育厅科研基金项目(2019J0829)。

摘　　要：针对传统图像复原算法对噪声强度较大的模糊图像进行复原时,容易出现图像失真的问题,研究结合经典维纳滤波算法和空间滤波算法,提出一种组合维纳滤波算法的噪声模糊图像复原方法。首先,采用空间滤波算法减弱模糊图像的噪声;然后,采用维纳滤波算法对去噪后的模糊图像进行复原;最后,通过仿真实验对组合维纳滤波算法进行验证。结果表明,相较于单一的维纳滤波算法,所研究组合维纳滤波算法在高斯噪声、椒盐噪声、高斯-椒盐噪声的条件下,PSNR、SNR、ISNR值均提高了0.5 dB以上,具有一定的优越性,可提高复原图像的质量,得到清晰且轮廓分明的复原图像。In order to solve the problem of image distortion when traditional image restoration algorithm is used to recover the fuzzy image with high noise intensity,a new method of noise fuzzy image restoration is proposed by combining the classical Wiener filtering algorithm and spatial filtering algorithm.Firstly,the spatial filtering algorithm is used to reduce the noise of the fuzzy image;then,the Wiener filter algorithm is used to recover the denoised fuzzy image;finally,the combined Wiener filter algorithm is veri-fied by simulation experiments.The results show that compared with the single Wiener filter algorithm,the PSNR,SNR and IS-NR values are all increased by more than 0.5dB under the condition of Gaussian noise,salt pepper noise and Gaussian salt pepper noise.The proposed method has some advantages,which can improve the quality of the restored image and get clear and clear contour restoration image.

关 键 词：图像复原 维纳算法 空间滤波 均值滤波 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]