【摘要】 以3种前人所提出的重磁边界识别方法与笔者提出的归一化偏差法作为技术手段(其中归一化偏差法对于数据噪声有着较好的抵抗力)能够较为准确全面地识别边界位置.在分析对比方法原理及应用特性的基础上,通过理论模型试验证明了方法的可靠性和实用性;并针对断裂和地质体边界深部倾向问题,开展了多方法联合及多尺度识别手段试验,获取了更多地质构造及边界的信息.通过将该技术应用于克拉玛依后山地区实际资料的处理解释之中,获得了研究区域内剖面与平面上的断裂和岩体的位置分布及它们的深部倾向信息,并对主要断裂与岩体三维空间特征进行了描述,为该地区三维地质填图和三维地质模型构建提供了重要依据.更多还原

【Abstract】 In this paper,four edge-detection methods including normalized deviation(ND)method,a new one proposed by us,are studied for their applications in 3-D mapping.The ND method is featured with low sensitivity to the noise,high accuracy and detailed results.On the basis of the theoretical framework of these methods,we tested their accuracy with synthetic models and gained more information of the geological boundaries by the combination of different methods and edge detecting on multiscale,which shows good performance.Then we used them in field data of the Karamay back mountain area to get the distribution of faults and rock masses,strikes and depths.The spatial features of some major faults and rock masses are described,which delivers useful information for the 3-D geological mapping of the study area.更多还原