摘要
在分析了硬域值、软域值小波去噪算法存在问题的基础上,提出了一个新的域值函数,该阈值函数在整个定义域内统一定义,表达式简单易于计算,同软阈值函数一样具有连续性,且是高阶可导的,便于进行各种数学处理。与硬软域值去噪方法相比,该改进的新阈值函数通过调节参数来调节阈值化小波系数与原始小波系数之间的恒定偏差以及过渡区内曲线的平滑性,弥补了硬阈值法不连续性和软阈值法具有偏差的不足。仿真结果表明:改进后的小波阈值函数能够得到较小的均方误差,并且去噪后的图像主观视觉效果和峰值信噪比均比传统算法优越。
On the basis of analysis of the problem of the traditional threshold function, the hard threshold function and the soft threshold function, a new threshold function is proposed. Compared with the traditional thresholding functions,it is more simple in expression, as continuous as the soft-thresholding function, and has a higher order derivative which makes some !rinds of mathematical disposals convenient. With this threshold function, through changing the value of the parameters, the constant deviation between threshold wavelet coefficient and original wavelet coefficient can be adjusted and the curve in transit area smooth is made, which overcomes the disadvantages of non-continuity in hardware threshold method and the deviation in soft threshold method. The experimental results show that the proposed method presents better MSE perfofinance and has both better subjective visual effect and the PSNR than the traditional approaches.
出处
《传感器与微系统》
CSCD
北大核心
2011年第9期76-78,81,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(61001170)
关键词
小波去噪
阈值函数
均方误差
信噪比
wavelet denoising
threshold function
mean square error（MSE）
signal noise ratio（SNR）
