基于图像工程的车载桥梁安全检查平台的探索
发布时间：2018-07-16 14:21
【摘要】：本文依托导师周志祥教授承担的重庆市自然科学基金课题：桥梁面相感知方法与安全评价理论研究（项目编号：cstc2012jjB0118），在查阅国内外研究现状的基础上提出并初步研发了“车载桥梁安全检查平台”，涉及到桥梁工程、机械工程、图像工程、计算机科学等学科，其主要研究内容和结论如下： ①通过分析桥梁安全检查的部位和内容，明确了车载桥梁安全检查平台的功能定位与检查技术指标，研究了结构的总体布置及各组成结构之间的衔接策略。车载桥梁安全检查平台是基于图像工程的新型桥梁健康检查设备，由臂架结构等机械装置和可视化桥梁健康检测系统组成，适用于量大面广的常规混凝土梁桥和拱桥的安全检查。 ②借助现有混凝土泵车蛇形臂的工作原理，构造了检测平台的臂架结构，其与底盘的连接采用一种可360°旋转的支撑架安装座；进行了二类底盘的比选及具体参数的确定；为了降低工作过程中机械部分振动对图像采集的不利影响，，本文提出设置液压缸液体弹簧、平衡轮和三级刚度螺旋弹簧悬架系统的措施，并利用MATLAB中的Simulink组件对底盘三级刚度螺旋弹簧悬架系统进行了仿真分析，分析结果表明该系统能有效降低平台工作时的自身振动。 ③为了实现对采集到的图像信息进行病害识别，提出自定义广义结构元与扩展形态学滤波相结合的方法在桥梁病害信息识别上的应用；初步开发了基于C#语言的可视化桥梁健康检测系统；提出矢量化存储和模拟重现的方法来降低数据的存储量。 ④通过图像采集现场试验中图像采集设备对桥梁底部病害信息的采集，验证了图像采集的效果可以满足桥梁安全检查的要求，能够识别桥梁结构表面的具体情况；通过基于桥梁表观信息的三维建模技术，对采集到的图像进行拼接和重组。通过实物模型图像动态采集与处理试验，进行了图像的静态采集和动态采集的对比分析，分析得出动态采集效果相对较差，但并不妨碍对图像的分析处理；通过控制检测速度和采用更高的帧率可以改善图像动态采集的效果。
[Abstract]:This paper relies on the subject of Chongqing Natural Science Foundation: the Theory of Bridge Phase perception and Safety Evaluation (Project No.: cstc2012jjB0118), which is undertaken by Professor Zhou Zhixiang, and puts forward and preliminarily studies it on the basis of consulting the current research situation at home and abroad. Issued "vehicle bridge safety inspection platform", Related to bridge engineering, mechanical engineering, image engineering, computer science and other disciplines, its main research contents and conclusions are as follows: 1 by analyzing the location and content of bridge safety inspection, In this paper, the function orientation and inspection technical index of the vehicle bridge safety inspection platform are defined, and the general arrangement of the structure and the linking strategy among the components are studied. The bridge safety inspection platform is a new type of bridge health inspection equipment based on image engineering. It is composed of mechanical devices such as arm structure and visual bridge health detection system. It is suitable for the safety inspection of conventional concrete beam bridges and arch bridges with a wide range of measures. 2 with the help of the working principle of the snake-shaped arm of the existing concrete pump vehicle, the jib structure of the detection platform is constructed. The connection between the chassis and the chassis adopts a 360 掳rotatable support frame mounting base. The comparison and selection of the second type chassis and the determination of the specific parameters are carried out. In order to reduce the adverse effect of mechanical vibration on image acquisition during work, In this paper, the measures of setting up hydraulic cylinder liquid spring, balancing wheel and three-stage stiffness spiral spring suspension system are put forward, and the simulation analysis of chassis three-stage stiffness spiral spring suspension system is carried out by using Simulink module in MATLAB. The analysis results show that the system can effectively reduce the vibration of the platform. 3 in order to realize the disease identification of the collected image information, The application of self-defined generalized structure element and extended morphological filter in bridge disease information identification is presented, and a visual bridge health detection system based on C # language is developed. The method of vectorized storage and simulated reproduction is proposed to reduce the storage capacity of data. 4 through image acquisition field test, the information of bridge bottom disease is collected by image acquisition equipment. It is verified that the effect of image acquisition can meet the requirements of bridge safety inspection, and can identify the concrete situation of bridge structure surface; through the 3D modeling technology based on the apparent information of bridge, the collected images can be stitched and reorganized. Through the dynamic acquisition and processing experiment of the real object model image, the contrast analysis of the static and dynamic image acquisition is carried out. The result shows that the dynamic acquisition effect is relatively poor, but it does not hinder the analysis and processing of the image. The effect of dynamic image acquisition can be improved by controlling detection speed and adopting higher frame rate.
【学位授予单位】：重庆交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U446
[Abstract]:This paper relies on the subject of Chongqing Natural Science Foundation: the Theory of Bridge Phase perception and Safety Evaluation (Project No.: cstc2012jjB0118), which is undertaken by Professor Zhou Zhixiang, and puts forward and preliminarily studies it on the basis of consulting the current research situation at home and abroad. Issued "vehicle bridge safety inspection platform", Related to bridge engineering, mechanical engineering, image engineering, computer science and other disciplines, its main research contents and conclusions are as follows: 1 by analyzing the location and content of bridge safety inspection, In this paper, the function orientation and inspection technical index of the vehicle bridge safety inspection platform are defined, and the general arrangement of the structure and the linking strategy among the components are studied. The bridge safety inspection platform is a new type of bridge health inspection equipment based on image engineering. It is composed of mechanical devices such as arm structure and visual bridge health detection system. It is suitable for the safety inspection of conventional concrete beam bridges and arch bridges with a wide range of measures. 2 with the help of the working principle of the snake-shaped arm of the existing concrete pump vehicle, the jib structure of the detection platform is constructed. The connection between the chassis and the chassis adopts a 360 掳rotatable support frame mounting base. The comparison and selection of the second type chassis and the determination of the specific parameters are carried out. In order to reduce the adverse effect of mechanical vibration on image acquisition during work, In this paper, the measures of setting up hydraulic cylinder liquid spring, balancing wheel and three-stage stiffness spiral spring suspension system are put forward, and the simulation analysis of chassis three-stage stiffness spiral spring suspension system is carried out by using Simulink module in MATLAB. The analysis results show that the system can effectively reduce the vibration of the platform. 3 in order to realize the disease identification of the collected image information, The application of self-defined generalized structure element and extended morphological filter in bridge disease information identification is presented, and a visual bridge health detection system based on C # language is developed. The method of vectorized storage and simulated reproduction is proposed to reduce the storage capacity of data. 4 through image acquisition field test, the information of bridge bottom disease is collected by image acquisition equipment. It is verified that the effect of image acquisition can meet the requirements of bridge safety inspection, and can identify the concrete situation of bridge structure surface; through the 3D modeling technology based on the apparent information of bridge, the collected images can be stitched and reorganized. Through the dynamic acquisition and processing experiment of the real object model image, the contrast analysis of the static and dynamic image acquisition is carried out. The result shows that the dynamic acquisition effect is relatively poor, but it does not hinder the analysis and processing of the image. The effect of dynamic image acquisition can be improved by controlling detection speed and adopting higher frame rate.
【学位授予单位】：重庆交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U446
【参考文献】
相关期刊论文 前10条
1 冯霏;刘杨;李凌轩;闻邦椿;;汽车非线性悬架最优控制的研究[J];中国工程机械学报;2010年02期
2 李军;李淑慧;;智能化桥梁检测车关键技术研究[J];公路交通科技(应用技术版);2011年05期
3 王华,朱宁,王祁;公路路面分形纹理特征分析和分类[J];哈尔滨工业大学学报;2005年06期
4 蒋立辉,耿蒙,赵春晖;基于广义形态滤波和模糊逻辑的散斑噪声抑制[J];红外与激光工程;2005年01期
5 杜文靖;朱少辉;刘玉臣;;高等级沥青公路路面裂缝图像处理技术[J];建设机械技术与管理;2006年03期
6 李建松;赵静一;王昕煜;李洪涛;;自行式桥梁检测车行走驱动系统特性分析[J];机床与液压;2010年10期
7 韩宏,杨静宇;神经网络分类器的组合[J];计算机研究与发展;2000年12期
8 张方,钱永久,索清辉;桥梁健康检测及其智能系统[J];河南理工大学学报(自然科学版);2005年02期
9 王s
本文编号：2126663
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2126663.html
