Bridge cracks present important information for the safety and the durability evaluationof the concrete bridge structure.It is of great significance to detect and analyze cracks for themaintenance of concrete bridges,and the demand has increased rapidly every year.However,the current methods for inspecting cracks in concrete bridges are still conventional,dangerous,qualitative,subjective and time-consuming due to relying mainly on the human vision.Inaddition,the complexity of the bridge structure may bring great difficulty to the human visioninspection procedure.To make the crack inspection more effective and safer,we develop a crackdetection robot system which is composed by a specially designed truck with a huge and flexiblemechanical arm and a machine vision system with various sensors.The developed system can automatically acquirethe sensor data for bridge bottom surfaces,and then detect and locate the cracks.Experiments show that the performance of our system is better than state-of-the-art systems,which is applicable in practice.%桥梁裂缝呈现出混凝土桥梁安全性、耐久性等重要信息.裂缝检测和分析对于桥梁的养护具有非常重要的意义,并且近年来这一需求日益激增.然而,目前桥梁检测手段依然是传统、危险、定性、客观和耗时的目视检测方法为主.此外,对于具有复杂结构的桥梁,传统的目视检测方法实现难度较大.为了使桥梁测更加高效和安全,本文提出一种用于桥梁检测的智能系统,该系统由特殊设计的安装了一个多自由度机械臂的移动平台和装备多种传感器的视觉系统组成.该系统可以自动地采集桥梁底面数据,并能识别和定位裂缝.实验验证显示,该系统性能明显优于目前已有系统,具有较强的实用性.
展开▼
