基于拟合误差消除的探地雷达图像鲁棒双曲线识别模型  

A Robust Hyperbola Recognition Model with Fitting-Errors-Based Eliminating in GPR B-scan Image

作　　者：兰天 赵毅[1,2] 陈宏畅 龚俊波 王长军 王健[3] 杨小鹏 LAN Tian;ZHAO Yi;CHEN Hongchang;GONG Junbo;WANG Changjun;WANG Jian;YANG Xiaopeng(Radar Research Lab,School of Information and Electronics,Beijing Institute of Technology,Beijing 100081,China;Beijing Institute of Technology Chongqing Innovation Center,Chongqing 401120,China;Beijing Building Research Institute Corporation Limited of CSCEC,Beijing 100070,China)

机构地区：[1]北京理工大学信息与电子学院雷达技术研究所,北京100081 [2]北京理工大学重庆创新中心,重庆401120 [3]北京中建建筑科学研究院有限公司,北京100070

出　　处：《信号处理》2023年第9期1699-1710,共12页Journal of Signal Processing

基　　金：国家自然科学基金项目(61860206012,61901441);重庆市自然基金项目(cstc2020jcyj-msxmX0768,cstc2021jcyj-msxmX0339);中建集团科技研发子课题(CSCEC-2020-Z-7-5)。

摘　　要：作为一种无损探测工具,探地雷达已广泛应用于各种场景,但从探地雷达B扫描图像中自动提取信息仍然具有挑战性。本文提出了一种从探地雷达B扫描图像中自动识别和拟合双曲线的鲁棒集成模型。首先,实现了由均值对消、基于梯度的自适应阈值算法和开闭操作组成的图像预处理方法,其中,均值对消抑制杂波和噪声,基于梯度的自适应阈值算法可以将B扫描图像转换为二值图像,然后通过开闭操作去除离散的噪声点;接下来,通过下开口扫描聚类算法识别具有向下开口的点集;之后,利用基于代数距离的双曲线拟合算法对这些点集进行直接拟合;最后,根据这些点集的拟合结果,基于拟合误差的剔除方法去除不具有完全双曲特征的下开口点集,从而实现B扫描图像中所有双曲点集的识别和拟合。由上述方法组成的集成模型能够自动、稳健地从探地雷达B扫描图像中提取信息。在合成数据集和实测数据集上的实验表明了所提出模型方法的有效性。As a nondestructive tool,ground-penetrating radar(GPR)has been widely used for the investigation of the sub⁃surface,but it is challenging to automatically extract information from GPR B-scan images.In this paper,a robust inte⁃grated model for automatically recognizing and fitting the hyperbolae from GPR B-scan images is proposed,which can eliminate non-hyperbolic clusters.Firstly,the preprocessing method which consists of the mean subtraction operation,the adaptive thresholding algorithm based on gradient,and the opening and closing operations is implemented.The mean sub⁃traction operation is utilized to suppress clutter and noise.And the adaptive thresholding algorithm based on gradient could transform the B-scan image to the binary image.Then the opening and closing operations remove discrete noise points.Next,point clusters with downward-opening are identified by open-scan-clustering algorithm(OSCA).After that,these point clusters are directly fitted by hyperbola fitting algorithm based on algebraic distance.Finally,based on the fitting re⁃sults of these point clusters,the fitting-errors-based eliminating(FEE)method removes downward-opening point clusters without complete hyperbolic feature,thus all hyperbolic point clusters in the B-scan image could be recognized and fitted.This integrated model consisting of methods above can automatically and robustly extract information from GPR B-scan im⁃ages.The experiments on synthetic and real datasets indicate the effectiveness of the proposed integrated model.

关 键 词：探地雷达 基于拟合误差消除方法 数据处理 图像处理 识别与拟合 

分 类 号：TN959.71[电子电信—信号与信息处理]