摘要
为了克服硬阈值函数不连续,软阈值函数中估计小波系数与分解小波系数之间存在着恒定偏差的缺陷,更好地改进滤波效果,提高去噪质量,须采用一种新的阈值处理策略。新阈值函数表达式简单易于计算,克服了硬软阈值的缺点。实验结果表明,该方法可以有效地去除白噪声干扰,无论是在视觉效果上还是在信噪比和均方误差定量指标上均明显优于常用的软、硬阈值及Matlab的阈值优化算法,充分体现出其优越性。
To overcome the defect of discontinuousness of hard threshold function and the constant deviation of estimated wavelet coef -ficients and descomposed wavelet coefficients of soft threshold function , and better improve the filtering result and enhance the denois-ing quality, a new threshold management strategy is adopted .The new threshold function expression is simple and easy to calculate , which overcomes the shortcomings of hard and soft threshold function .The result of this experiment displays that this method can wipe out the white noise efficiently .It is superior to common hard and soft threshold and matlab threshold optimization algorithm whether in visual effect or noise ratio and mean squared error , which fully reflects the superiority of wavelet threshold denoising method .
出处
《测绘与空间地理信息》
2014年第4期205-207,共3页
Geomatics & Spatial Information Technology
关键词
小波变换
小波阈值去噪
阈值函数
均方误差
信噪比
wavelet transformation
wavelet threshold denoising
threshold function
MSE
SNR
