图像信息熵约束的浅地层层界划分方法

发布时间：2018-07-02 23:55

本文选题：浅地层剖面图像 + 浅地层层界及提取　；参考：《哈尔滨工业大学学报》2017年08期

【摘要】：为实现快速、精确、自动化、智能化的海底浅地层层界提取,克服传统浅地层层界在复杂海洋环境下提取时的低效、模糊、主观性等缺点,提出一种基于图像信息熵约束的浅地层层界划分方法.首先,将浅剖图像分割为不同区块;然后,在不同区块计算信息熵,并结合钻孔数据,建立信息熵与显著性参数关系模型;最后,据此模型对整个浅剖图像进行层界划分.研究表明,该方法克服了现有方法的不足,实现了浅地层剖面层界的自适应、准确划分,试验中取得了与钻孔层界深度、厚度同量级的精度.由此可知采用图像信息熵约束进行层界提取,可以实现浅地层层界提取的自动化与智能化.
[Abstract]:In order to realize the fast, accurate, automatic and intelligent extraction of the shallow layer boundaries of the sea floor, and overcome the shortcomings of the traditional shallow layer boundaries extraction in complex marine environment, such as inefficiency, fuzziness, subjectivity, etc. A method of dividing shallow layer boundaries based on entropy constraint of image information is proposed. Firstly, the shallow-cut image is divided into different blocks; then, the information entropy is calculated in different blocks and the relationship between information entropy and salient parameters is established by combining the borehole data. Finally, the whole shallow profile image is divided into layers according to the model. The research shows that the method overcomes the shortcomings of the existing methods and realizes the self-adaptation and accurate division of the stratigraphic boundary of the shallow strata. The accuracy of the experiment is the same as the depth and thickness of the borehole boundary. It can be concluded that using the restriction of image information entropy to extract the layer boundary can realize the automation and intelligence of the shallow layer boundary extraction.
【作者单位】：武汉大学测绘学院;武汉大学动力与机械学院;
【基金】：国家自然科学基金(41376109,41176068,41576107)
【分类号】：P229;TP391.41

【相似文献】