2D地震小波变换和自组织网络联合边界检测

Edge detection jointly with 2-D seismic wavelet transform and self organizing network

1. 石油大学地球资源与信息学院,山东东营257062;2. 中国石化胜利油田物探研究院,山东东营255075

1. Faulty of Earth Resource and Information Technology, University of Petroleum, Dongying, 257062, China;2 .Geophysical Research Institute, SINOPEC Shengli Oilfield Company Limited, Dongying 255075, China

阐述了利用信号二维小波变换的特征点进行边界检测的理论方法及可行性, 以及地震信号二维小波变换的特征点的提取计算。讨论了不同尺度下二维小波变换特征点的性质、特点, 研究了利用自组织网络对提取计算的二维小波变换的特征点进行竞争学习、聚类。并根据所研究的问题, 设计了相应的算法, 通过实际资料的应用, 取得了较好的效果。

This paper discussed the feasibility of edge detection on seismic data with 2-D wavelet transform. Method for calculation of the characteristic points of 2-D wavelet transform was presented. The characteristic points were then clustered by self organizing network. Desired results have been achieved by applying the method to real data.