基于图像纹理分析的海域磁异常分区

Marine Magnetic Anomaly Partition Based on Image Texture Analysis

摘要:

摘 要：将图像纹理分析技术应用于海洋磁异常资料的分区，采用Gabor滤波器和灰度共生矩阵相结合的图像纹理分析方法，提取海洋磁异常的纹理特征信息，通过OKMS聚类算法实现不同纹理特征区域磁异常场的分区。模型试验结果表明，利用Gabor滤波器和灰度共生矩阵相结合的图像纹理分析方法可有效提取磁异常的纹理特征信息，对磁异常纹理特征信息的聚类分区取得较好效果。实际应用于加罗林海域磁异常的分区结果显示，加罗林海域可分为11个磁异常区块，分区结果与条带磁异常区和岩浆活动区对应较好，该方法为海洋磁异常的分区解释提供了参考依据。

Abstract:

Abstract: An image texture analysis method integrated Gabor filter and Gray-level co-occurrence matrix is used to extract the texture features of marine magnetic anomalies. Using OKMS clustering algorithm realized magnetic anomaly field partition of different texture features .The model test and actual application results show that image texture analysis method integrated Gabor filter and Gray-level co-occurrence matrix can effectively extract the texture characteristics information of magnetic anomalies. The partition results of Caroline Marine magnetic anomaly show that Caroline plate magnetic anomaly can be divided into 11 areas, partition results coincide well with strip magnetic anomaly areas and magmatic activity areas,and this method provides a reference for processing and interpretation of magnetic anomaly