数字钻孔图像兴趣区域分割与岩体结构面特征识别

针对现有数字钻孔图像分析技术的不足，提出新分析方案以实现数字钻孔摄像技术（BCT）所采集钻孔内壁图像的自动化、定量化结构面检测、识别与分析.首先，对数字钻孔图像进行预处理，设计特征信号DH，检测并获取兴趣区域；然后，使用低精度Hough变换快速检测结构面兴趣区域内的结构面分布特征，利用聚类算法分离出单一结构面后，针对单一结构面进行亚像素级Hough变换，以获取结构面的正弦参数；最后，根据结构面正弦参数，计算出结构面的倾向、倾角等信息.通过对如美水电站左坝肩数字钻孔图像实例进行分析，利用本算法完成图像中结构面的自动分割与特征识别，成功获取其几何信息，并与传统人工辅助方案结果进行对比，验证了该算法的可靠性. 

Segmentation of Region of Interest and Identification of Rock Discontinuities in Digital Borehole Images

Aiming at the limitations of previous Digital Borehole Image Analysis Technology, a new analysis scheme to realize automatic quantitative analysis of borehole inner wall images collected by Borehole Camera Technology (BCT) was developed. First, feature signal DH was designed to detect the pre-treated image and obtain Region of Interest (ROI) automatically. Second, low accuracy Hough transform was used to detect the distribution of discontinuities in the ROI rapidly. After separating discontinuities with clustering algorithm, the sub-pixel Hough transform was carried out for single discontinuity to obtain its sine parameters. Finally, the dip direction and dip of each single discontinuity were calculated using its sine parameters. In the case study, the automatic segmentation and feature recognition of the discontinuities in the digital images from the left bank of Rumei hydropower station were completed using proposed method, and the geometric data of discontinuities matched with the results of traditional artificial assistance method, verifying the reliability of proposed method.