提高光电成像挠度系统的识别精度  

Improving Recognition Accuracy of Optoelectronic Imaging Deflection System

作　　者：胡顺仁[1,2] 

机构地区：[1]重庆理工大学电子信息与自动化学院,重庆400050 [2]重庆大学光电技术及系统教育部重点实验室,重庆400044

出　　处：《光电工程》2014年第2期1-5,共5页Opto-Electronic Engineering

基　　金：国家863计划项目(2006AA04Z433);国家科技支撑计划项目(2007BAE15B04);重庆市科技攻关重大专项(7289)

摘　　要：光电成像挠度系统在桥梁结构状态健康监测系统工程实践中由于受到雾、强光等天气因素干扰会影响识别精度。针对该问题,本文在分析了光电成像挠度测量系统组成和成像识别原理基础上,深入研究了各种不利的天气因素干扰对挠度测量系统成像识别精度的影响,并在图像增强基础上提出一种投影差分算法来提高在不利天气条件下光电成像识别精度,该算法通过对比实验(雾天:该方法的均方差为1.632 231,传统方法均方误差为16.534 08;强光:该方法的均方差为2.182 247,传统方法均方误差为23.018 13)验证了该方法在不利天气条件下相比传统方法将识别精度从厘米级提高到毫米级,大大增强了光电成像挠度系统的可靠性,并在重庆马桑溪长江大桥、重庆高家花园嘉陵江大桥等桥梁结构状态监测系统的光电成像挠度测量系统工程实践中得以成功检验。Because of high accuracy and wide range, the advanced optoelectronic imaging deflection system has become the important device of the Bridge Structural Health Monitoring System （BSHMS）, and has been applied to several bridges. However, the recognition accuracy of the optoelectronic imaging deflection system has greatly descended for fog and sunshine in practice. By analyzing the principle of image-based deflection measurement, a conclusion was drawn that small error in light source target in image recognition caused larger error in deflection because of amplification of optical measurement system. A novel method based on projection and differential is proposed for recognition of light source target. Compared with conventional method （its Mean-Square Error （MSE） is 16.534 08＆lt;fog＆gt;and 23.018 13＆lt;sunny＆gt;）, the proposed approach （its MSE is 1.632 231＆lt;fog＆gt;and 2.182 247＆lt;sunny＆gt;） is more accurate and accords with practice. Experiment results verify the effectiveness of the designed method and theoretical discussions.

关 键 词：光电成像 挠度 识别精度 投影差分 图像增强 

分 类 号：U446.2[建筑科学—桥梁与隧道工程] TP274.2[交通运输工程—道路与铁道工程]