...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Earth, planets and space: EPS >Processing the Bouguer anomaly map of Biga and the surrounding area by the cellular neural network: application to the southwestern Marmara region
【24h】

Processing the Bouguer anomaly map of Biga and the surrounding area by the cellular neural network: application to the southwestern Marmara region

机译：利用细胞神经网络处理Biga及其周边地区的布格异常图：在西南马尔马拉地区的应用

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

An image processing technique called the cellular neural network (CNN) approach is used in this study to locate geological features giving rise to gravity anomalies such as faults or the boundary of two geologic zones. CNN is a stochastic image processing technique based on template optimization using the neighborhood relationships of cells. These cells can be characterized by a functional block diagram that is typical of neural network theory. The functionality of CNN is described in its entirety by a number of small matrices (A, B and I) called the cloning template. CNN can also be considered to be a nonlinear convolution of these matrices. This template describes the strength of the nearest neighbor interconnections in the network. The recurrent perceptron learning algorithm (RPLA) is used in optimization of cloning template. The CNN and standard Canny algorithms were first tested on two sets of synthetic gravity data with the aim of checking the reliability of the proposed approach. The CNN method was compared with classical derivative techniques by applying the cross-correlation method (CC) to the same anomaly map as this latter approach can detect some features that are difficult to identify on the Bouguer anomaly maps. This approach was then applied to the Bouguer anomaly map of Biga and its surrounding area, in Turkey. Structural features in the area between Bandirma, Biga, Yenice and Gonen in the southwest Marmara region are investigated by applying the CNN and CC to the Bouguer anomaly map. Faults identified by these algorithms are generally in accordance with previously mapped surface faults. These examples show that the geologic boundaries can be detected from Bouguer anomaly maps using the cloning template approach. A visual evaluation of the outputs of the CNN and CC approaches is carried out, and the results are compared with each other. This approach provides quantitative solutions based on just a few assumptions, which makes the method more powerful than the classical methods.
机译：在这项研究中，使用了一种称为细胞神经网络（CNN）方法的图像处理技术来定位引起重力异常的地质特征，例如断层或两个地质区域的边界。 CNN是一种基于模板优化的随机图像处理技术，该模板使用单元的邻域关系。这些单元可以通过神经网络理论中典型的功能框图来表征。 CNN的功能通过称为克隆模板的许多小矩阵（A，B和I）完整描述。 CNN也可以被认为是这些矩阵的非线性卷积。该模板描述了网络中最近邻居的强度。循环感知器学习算法（RPLA）用于克隆模板的优化。 CNN和标准Canny算法首先在两组合成重力数据上进行了测试，目的是检验所提出方法的可靠性。通过将互相关方法（CC）应用于相同的异常图，将CNN方法与经典派生技术进行了比较，因为后者可以检测出在布格异常图上难以识别的某些特征。然后将此方法应用于土耳其Biga及其周边地区的布格异常地图。通过将CNN和CC应用于布格异常图，研究了西南马尔马拉地区班迪尔马，比加，耶尼丝和戈嫩之间的区域的结构特征。这些算法确定的故障通常与先前映射的表面故障一致。这些示例表明，可以使用克隆模板方法从布格异常图中检测出地质边界。对CNN和CC方法的输出进行视觉评估，并将结果进行比较。这种方法仅基于一些假设就可以提供定量解决方案，这使该方法比经典方法更强大。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号