基于自适应阈值Canny算法的裂缝检测方法研究
|
摘要
随着科学技术的不断发展,桥梁、隧道、高速公路等土木工程设施开始大量出现在人们的生活之中。它们在给予人们方便之余,也存在着大量的安全隐患,而由于土木工程体量、处在地自然环境等因素的限制,现行的人工检测方法很难及时地对土木工程进行结构健康监测,数字图像处理成为解决这个问题的首选之一。文中首先对现行的Canny裂缝检测算法进行了详细的介绍,并针对其只能人工选取阈值的缺点进行了改进,结合Harris特征检测算法和图像中各像素点的梯度值,提出了一种自适应阈值的Canny检测算法;然后,结合自身的学习,在现有的两种裂缝评价指标裂缝宽度和长度之外,引入了裂缝的横向位移和旋转角度,构建了新的裂缝安全评估指标。通过实验对所提出的算法进行了验证。
With the continuous development of science and technology,bridges,tunnels,highways and other civil engineering facilities have begun to appear in people's lives. They give people convenience,but there are also a lot of potential safety hazards. Due to the limits of civil engineering volume,the natural environment and other factors,the current method of manual detection can not be very good and timely for structural health monitoring of civil engineering,thus digital image processing has become one of the first to solve the problem. This paper firstly introduces the current crack detection algorithm of Canny,and makes some improvement for the only artificial selection of threshold. Then,in addition to Harris corner detection and the crack width and length of the existing two kinds of fracture evaluation index,the new fracture safety assessment index is introduced,which is introduced by the transverse displacement and rotation angle of the crack. The validity of the proposed algorithm is verified by experiments.
With the continuous development of science and technology,bridges,tunnels,highways and other civil engineering facilities have begun to appear in people's lives. They give people convenience,but there are also a lot of potential safety hazards. Due to the limits of civil engineering volume,the natural environment and other factors,the current method of manual detection can not be very good and timely for structural health monitoring of civil engineering,thus digital image processing has become one of the first to solve the problem. This paper firstly introduces the current crack detection algorithm of Canny,and makes some improvement for the only artificial selection of threshold. Then,in addition to Harris corner detection and the crack width and length of the existing two kinds of fracture evaluation index,the new fracture safety assessment index is introduced,which is introduced by the transverse displacement and rotation angle of the crack. The validity of the proposed algorithm is verified by experiments.
引文
[1]邹飞,李海波,周青春,等.基于数字图像灰度相关性的类岩石材料损伤分形特征研究[J].岩土力学,2012,33(3):731-737.
[2]NISTER D.Preemptive RANSAC for live structure and motion estimation[J].Machine Vision and Applications,2005,16(5):321-329.
[3]邹飞,李海波,周青春,等.岩石节理倾角和间距对隧道掘进机破岩特性影响的试验研究[J].岩土力学,2012,33(6):1640-1646.
[4]李卉.数据可视化技术在物联网监控系统中的研究与应用[D].北京:北京邮电大学,2013.
[5]ROSTEN E,DRUMMOND T.Machine learning for high-speed corner detection[M].Computer Vision-ECCV 2006.Springer Berlin Heidelberg,2006:430-443.
[6]徐浩.市政桥梁安全监控网络系统的研究[D].合肥:合肥工业大学,2005.
[7]贺强,晏立.基于LOG和Canny算子的边缘检测算法[J].计算机工程,2011,37(3):210-212.
[8]陈宏希.基于边缘保持平滑滤波的Canny算子边缘检测[J].兰州交通大学学报,2006,25(1):86-90.
[9]王涛,徐娅萍,亢海龙.结合Canny算子的图像二值化方法[J].微型电脑应用,2010,26(2):4-7.
[10]丰晓霞.基于深度学习的图像识别算法研究[D].太原:太原理工大学,2015.
[2]NISTER D.Preemptive RANSAC for live structure and motion estimation[J].Machine Vision and Applications,2005,16(5):321-329.
[3]邹飞,李海波,周青春,等.岩石节理倾角和间距对隧道掘进机破岩特性影响的试验研究[J].岩土力学,2012,33(6):1640-1646.
[4]李卉.数据可视化技术在物联网监控系统中的研究与应用[D].北京:北京邮电大学,2013.
[5]ROSTEN E,DRUMMOND T.Machine learning for high-speed corner detection[M].Computer Vision-ECCV 2006.Springer Berlin Heidelberg,2006:430-443.
[6]徐浩.市政桥梁安全监控网络系统的研究[D].合肥:合肥工业大学,2005.
[7]贺强,晏立.基于LOG和Canny算子的边缘检测算法[J].计算机工程,2011,37(3):210-212.
[8]陈宏希.基于边缘保持平滑滤波的Canny算子边缘检测[J].兰州交通大学学报,2006,25(1):86-90.
[9]王涛,徐娅萍,亢海龙.结合Canny算子的图像二值化方法[J].微型电脑应用,2010,26(2):4-7.
[10]丰晓霞.基于深度学习的图像识别算法研究[D].太原:太原理工大学,2015.