探地雷达信号小波变换去噪  

Noise Elimination from Wavelet Transform by Using GPR Signal

作　　者：刘秀娟[1] 邓世坤[2] 徐保林 

机构地区：[1]南京大学海岸与海岛开发教育部重点实验室,南京210093 [2]中国地质大学地球物理与空间信息学院,武汉430074 [3]武汉科岛地理信息工程公司,武汉430021

出　　处：《工程勘察》2006年第10期66-71,共6页Geotechnical Investigation & Surveying

摘　　要：本文简述了小波变换去噪的方法原理,并用该方法对含有随机干扰的合成探地雷达记录及野外实测数据进行了处理。对理论模型的处理结果表明,相对于常规去噪方法来说,小波变换去噪法能更彻底去除噪声、完整保留信号细节。而地基强夯加固探地雷达剖面图像经小波去噪处理后,不同介质界面变得明显清晰,墩长及墩形较处理前得到突出,尤其是土石墩剖面,墩形及墩长较清晰可辨。理论模型及实测数据研究结果均表明,小波变换去噪法能有效改善探地雷达剖面质量,提高解释的准确性和精确性。The principle of the noise elimination with wavelet transform is briefly introduced and wavelet transform is used to process the synthetic signal of ground penetrating radar （GPR） with random noise and the field testing data. The results from the theoretical model processing show that the wavelet transform can eliminate more noise and reserve more signal detail compared to the traditional methods of noise elimination. The interface between different media becomes more legible and the figure of t!ae target is focused after the field data are processed with the wavelet transform. The results from the theoretical model and field datum processing with wavelet transform demonstrate that the wavelet transform can effectively improve the quality of GPR profile and increase the veracity and accuracy of the interpretation of the fieM GPR profile.

关 键 词：探地雷达 去噪 小波变换 

分 类 号：P631[天文地球—地质矿产勘探]