摘要
图像去噪问题一直是图像处理领域关注的问题之一。在图像去噪过程中,由于软阈值函数获得的去噪估计系数与原小波系数具有的恒定差,进而影响了重建图像的质量。为了有效地解决这一问题,在软阈值函数的基础上,提出了一种新的阈值函数用于图像的去噪重建。同时,新提出的阈值函数有效解决了硬阈值函数的不连续性问题。为了增加新阈值函数的灵活性,新阈值函数中添加了2个调节因子。实验结果表明:所提出的阈值函数优于经典的软、硬阈值函数,有效解决了软、硬阈值函数存在的缺陷,获得的重建图像质量和峰值信噪比有显著提高。
Image denoising is one of the widely discussed topics in the field of image processing.In image denoising,there is an invariable dispersion between the denoising estimated coefficients and the decomposed wavelet coefficients in the soft threshold function,which affects the quality of reconstructed image.To solve this problem,a new threshold function based on soft threshold function was proposed.At the same time,the new threshold function could solve the discontinuous of the hard threshold function.To increase the flexibility of the new threshold function,two adjusting factors were added.The experimental analysis shows that new threshold function is superior to the classical soft threshold and hard threshold,which effectively solves the defects of soft and hard threshold.The quality and peak signal to noise ratio of the reconstructed images obtained by the new threshold function have significant improvements.
作者
张杰
李银华
张焕龙
贺振东
史小平
ZHANG Jie;LI Yin-hua;ZHANG Huan-long;HE Zhen-dong;SHI Xiao-ping(College of Electrical and Information Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,China;School of Astronautics,Harbin Institute of Technology,Harbin 150008,China)
出处
《科学技术与工程》
北大核心
2020年第24期9918-9922,共5页
Science Technology and Engineering
基金
国家自然科学基金(61427809)
河南高等学校重点科研项目计划(20A510016)
郑州轻工业大学2019年众创空间孵化项目(2019ZCKJ231)。
关键词
图像去噪
软阈值函数
硬阈值函数
峰值信噪比
image denoising
soft threshold functionl
hard threshold function
peak signal to noise ratio
