基于图像处理的桥梁裂缝识别及特征匹配研究

Study on Bridge Crack Detection and Feature Matching Based on Image Processing

摘要: 裂缝作为混凝土桥梁最常见的病害之一，可以体现混凝土结构的损伤状况。针对混凝土裂缝图像采集过程中干扰因素较多、采用单一的固定参数进行图像降噪处理时存在着无法满足众多场景等问题，该文基于连通域原理，提出自适应阈值的裂缝降噪处理方法，通过利用裂缝的特征信息进行自适应调节图像降噪过程中的参数设定。此外，为了更好地评估桥梁的病害状况，本文提出基于Hu不变矩的裂缝特征图像匹配研究，进而可以实现对裂缝病害发展状态的跟踪及预测。通过实验验证表明，基于该方法的裂缝特征匹配具有可行性。

Abstract: As one of the most common diseases of concrete bridges, cracks can reflect the damaged condition of concrete structures. To address the problem that there are many interference factors in the process of concrete crack image acquisition, and a single fixed parameter for image noise reduction processing cannot meet many situations, based on the principle of connected domain, an adaptive threshold crack noise reduction processing method is proposed in this paper, which adaptively adjusts the parameter settings in the image noise reduction process by using the feature information of cracks. In addition, in order to better assess the condition of the bridge, a crack image matching study based on Hu-invariant moments is proposed in this paper, which in turn enables the tracking and prediction of the crack disease development status. The experimental verification shows that the crack feature matching based on this method is feasible.