Probabilistic robust hyperbola mixture model for interpreting ground penetrating radar data
Probabilistic robust hyperbola mixture model for interpreting ground penetrating radar data
复制标题
DOI：
10.1109/ijcnn.2010.5596298
复制
发表时间：
2010-07
期刊：
影响因子：
--
通讯作者：
Huanhuan Chen;A. Cohn
中科院分区：
文献类型：
--
作者：
Huanhuan Chen;A. Cohn
提出了一种基于分类期望最大化算法的概率稳健双曲线混合模型，并将该算法应用于探地雷达空间数据的解译。以往采用Hough变换或神经网络识别探地雷达双曲线的方法，由于其计算要求和神经网络方法难以获得足够合适的训练数据，不适合现场应用。该算法将一种基于正交距离的稳健双曲线拟合算法引入到概率混合模型中，可以实时识别探地雷达数据中的双曲线，并计算出埋地公用事业管道的深度和尺寸。双曲线的数目可以通过使用贝叶斯信息准则进行模型选择来确定。对合成/模拟数据和真实探地雷达数据的实验结果表明了该算法的有效性。
This paper proposes a probabilistic robust hyperbola mixture model based on a classification expectation maximization algorithm and applies this algorithm to Ground Penetrating Radar (GPR) spatial data interpretation. Previous work tackling this problem using the Hough transform or neural networks for identifying GPR hyperbolae are unsuitable for on-site applications owing to their computational demands and the difficulties of getting sufficient appropriate training data for neural network based approaches. By incorporating a robust hyperbola fitting algorithm based on orthogonal distance into the probabilistic mixture model, the proposed algorithm can identify the hyperbolae in GPR data in real time and also calculate the depth and the size of the buried utility pipes. The number of the hyperbolae can be determined by conducting model selection using a Bayesian information criterion. The experimental results on both the synthetic/simulated and real GPR data show the effectiveness of this algorithm.