摘要
对我国长江口和东海海区的典型声学地层剖面图所显示的图像特征的物理本质作了分析，归纳出９个方面的主要图像特征，并指出可以分别根据其中４个特征对９类可能存在的地层的地质属性进行分类和识别．在此基础上本文讨论了应用基于概率计算的Ｂａｙｅｓ图像分类器对声学地层剖面图进行地质解释的问题．
The paper analyzes the physical properties of pattern characteristics which are displayed on typical Acoustic Sea-bed Profiling Records (ASPAs) in the areas of Changjiang Estuary and the East China Sea. Nine pattern charac- teristics are summarized and it is shown that 9 geographical categories can be determined by 4 pattern characteristics. Based on the above analysis, a Bayes-based pattern characteristics classifier for interpretation of ASPRs is developed.
出处
《声学学报》
EI
CSCD
北大核心
2001年第4期365-371,共7页
Acta Acustica
基金
国家863计划青年基金项目(820-Q-09)
