结合路面深度影像梯度方向直方图和分水岭算法的裂缝检测

裂缝检测对于道路维护和管理具有重要作用.由于深度影像对路面油污、阴影等因素不敏感，近些年来基于深度影像的检测方法已成为路面裂缝检测新的研究方向之一.传统的激光扫描线方法没有顾及裂缝在整个空间分布的变异性、各向异性和全局性特征，无法有效检测横向、块状、网状等裂缝.针对以往算法的不足，提出一种结合梯度方向直方图和分水岭算法的路面裂缝检测方法.首先，通过梯度方向直方图算法提取路面深度影像的裂缝边缘强度和方向；然后，利用裂缝边缘方向改进传统分水岭算法，最终提取裂缝目标.实验结果表明，该方法不仅能够准确检测多种类型的裂缝目标，而且能识别裂缝破损程度.

Pavement crack detection fused HOG and watershed algorithm of range image

Pavement crack detection plays an important role in pavement maintaining and management. In recent years， pavement crack detection technique based on range image is a recent trend due to its ability of discriminating oil spills and shadows. Existing pavement crack detection methods didn't effectively detect transverse and reticular cracks， because these methods generally represented the geometry feature of crack using single laser scanning lines， which didn't take the effects of spatial variability， anisotropy and integrity into account. Aimed at the existing arithmetic insufficiency， the pavement crack detection fused histogram of oriented gradient and watershed algorithm was proposed. Firstly， crack edge location was detected using histogram of oriented gradient in road range image， and edge strength and orientation were obtained. Then， crack segmentation based on traditional watershed method was improved by edge orientation. Experiment results show that the proposed method is able to accurately detect different type of crack edge and identify the extent of the damage cracks simultaneously.