摘要
由于野外采集地震资料往往带有较多的随机噪声,给资料解释造成困难。针对小波阈值去噪的阈值选取通常需要对信号进行先验估计,带有较强猜测性,阈值选取难以获得最优结果。本文提出基于改进混沌果蝇优化的小波阈值法,将基于广义交叉验证(GCV)函数设定为阈值选取目标函数,在混沌果蝇优化算法中引入调节系数实现对该目标函数的迭代寻优,在无先验信息前提下,获取最优小波阈值。通过将本文算法用于合成地震记录和实际地震记录进行去噪处理,并对比常用小波阈值去噪算法,证明了本文算法的有效性。
It is usually difficult to interpret the seismic data collected in the field bearing a large amount of random noise. Although the wavelet threshold method can be used to remove such noise, it requires prior estimation of the signal, which is largely conjecture, hard to obtain optimal results. In this article, we propose a wavelet threshold denoising method based on the improved chaotic fruit fly optimization. This method selects the objective function based on the generalized cross validation （GCV） as the threshold selection function. In order to optimize this objective function, adjustment coef- ficient is introduced into the optimization algorithm of the chaotic fruit fly. Then the optimal wavelet threshold can be obtained without any prior informa- tion. By denoising tests of synthetic seismic records and measured seismic data and comparing with the commonly - used wavelet threshold denoising method, the effectiveness of this new method is proved.
出处
《地质与勘探》
CAS
CSCD
北大核心
2017年第4期765-772,共8页
Geology and Exploration
基金
国家自然科学基金项目(编号51304050)
东华理工大学核技术应用教育部工程研究中心开放基金(编号HJSJYB2015-13
HJSJYB2016-9
HJSJYB2016-1)
江西省自然科学基金(编号20161BBE53006)联合资助
关键词
地震信号
随机噪声
小波阈值法
混沌果蝇
seismic signal, random noise, wavelet threshold method, chaotic fruit fly
