基于分布式CCD/GPS的矿区沉陷灾害动态监测融合算法研究

项目名称： 基于分布式CCD/GPS的矿区沉陷灾害动态监测融合算法研究

项目编号： No.41272373

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 天文学、地球科学

项目作者： 丁安民

作者单位： 河南理工大学

项目金额： 85万元

中文摘要： 针对当前矿区沉陷灾害监测方法的局限性，本项目提出基于分布式CCD/GPS的矿区沉陷灾害动态监测方法。该方法具有连续、动态监测的功能，可及时预报严重的矿区沉陷灾害。实时的数据处理算法是该方法的核心技术。 以分形技术和Rodrigues为理论基础，本项目主要进行所提方法的数据处理算法研究。主要研究内容：根据矿区三维数字模型与预测的特征点信息，建立样本图像和整体图像的分形特征，进而结合基准点的GPS定位信息，进行基于分形特征与GPS定位信息融合的立体像对匹配算法研究；在立体像对匹配的基础上，进行基于双Rodrigues与GPS定位信息融合的立体快速定位算法研究；重点研究基于双Rodrigues与GPS定位信息融合的立体相对定向模型，以及基于双Rodrigues和GPS定位信息反演CCD绝对定向参数，实现监测系统的自动化修正。 该项目的研究可为矿区沉陷灾害动态监测提供新的技术手段和理论支持。

中文关键词： 分布式CCD；动态监测；罗德里格斯；融合算法；地质灾害监测预测

英文摘要： According to the disadvantages of traditional methods of mining subsidence, the dynamic monitoring method based on distributed CCD/GPS sensors for mining subsidence is proposed. This method, which can collect data dynamically and continuously, can forecast serious mining subsidence disasters in good time. Real-time algorithm is the key technology of this method. On the theories of fractal and Rodrigues, the fusion algorithm of dynamic monitoring for mining subsidence disasters by using distributed CCD/GPS sensors is proposed. The main contents of this project including: according to the DEM (Digital Elevation Model) of mining area and the information of feature points, the fractal signatures of sample image and the whole image will be built; And the stereo matching algorithm based-on fractal signatures and the position information provided by GPS will be studied; On the basis of stereo matching, the rapid position algorithm based-on bi-Rodrigues and the position information provided by GPS will be studied furthermore; the relative orientation algorithm based-on bi-Rodrigues and the position information provided by GPS will be studied chiefly. The deformation detection of base station and its elements inversion calculation of absolute orientation will be also studied chiefly. And the monitoring system can be revi

英文关键词： distributed CCD (Charge Coupled Device)；dynamic monitoring；Rodrigues；fusion algorithm；geological hazard monitoring and prediction