基于平衡区间多小波理论的图像去噪研究  

Study of image de-noising based on balanced and interval multi-wavelets

作　　者：周朴雄[1] 

机构地区：[1]华南理工大学电子商务学院,广东广州510006

出　　处：《计算机工程与设计》2008年第21期5513-5515,共3页Computer Engineering and Design

基　　金：广东省自然科学基金项目(07300647)

摘　　要：多小波在实际应用中需要对信号进行预滤波,这破坏了多小波的特性。提出了一种基于平衡区间多小波变换的综合闽值图像去噪方法。该方法通过对含噪图像进行区间多小波变换,一方面,克服了单小波变换中无法同时满足正交性和对称性的缺点,且因为区间多小波能够直接适应信号空间,所以不需要像普通小波信号处理那样进行信号的边界延;另一方面,因为平衡区间多小波的平衡性,因此在应用中不需要对图像进行预滤波。将综合阀值处理方法与平衡区间多小波变换相结合,根据多小波分解后的能量分布特性,在不同尺度的子带选择不同的最佳阈值,有效地提高了重构图像质量。实验结果表明,相对于非平衡多小波阈值处理方法,文中的去噪方法在输出信噪比扣主观视觉效果上都有明显改善。The pre-filter of signal is necessary in the application of multi-wavelet, which break characteristic of multi-wavelet. A synthesis threshold method based on the balanced and interval multi-wavelets is proposed. To apply the balanced and interval multi-wavelets transform to the noising image, on the hand, the method overcomes the limitation of single wavelets that can＇ t have orthogonality and symmetry simultaneously and haves no used for extending in boundary of signal; another hand, haves no used pre-filter for image. According to wavelet coefficients energy distribution in different sub-bands, different thresholds are obtained in different sub-bands to improve reconstructed image quality significantly. The experimental results show that compared with multi-wavelets methods, the method combining interval multi-wavelet transform and new threshold method greatly improves the PSNR value and subjective visual effects.

关 键 词：平衡区间多小波 图像去噪 阈值 预滤波 图像质量 

分 类 号：TN911.73[电子电信—通信与信息系统] TN918.1[电子电信—信息与通信工程]