摘要
针对两跨连续梁实验模型,通过采集连续梁振动视频,再将视频转化为数字图像的方法,运用多项式拟合法作为视频图像边缘检测技术中亚像素精度图像边缘识别的算法,对每幅图像进行亚像素边缘识别检测,获取各时刻梁边缘各点位置,经Matlab程序读取整个振动过程梁边缘数据,即可得到梁边缘各点的振动时间历程,进而通过DASP系统对测得数据进行模态分析,得到连续梁前二阶频率和振型。对比采用传统加速度传感器识别连续梁振动信号分析结果表明,两者模态参数识别结果具有一致性,基于视频图像边缘检测技术的连续梁振动测试是可行的。
Based on the experimental model of the two-span continuous beam,through the collection of video of vibration of the continuous beam,and by using the method of converting the video into digital image,the polynomial fitting method is used as the calculation method of the sub-pixel precision image edge recognition from the image edge detection technology to detect the sub-pixel edge for each picture,and obtain the position of the edge of the beam at each moment.Through acquiring the data of the entire beam edge vibration by Matlab program,the vibration time history of the beam edge can be acquired,and through the DASP system for the modal analysis of the measured data,the first two-order frequencies and vibration type of the continuous beam can be received.Compared with the traditional vibration signal analysis,the results show that the modal parameter identification results are consistent,and the continuous beam vibration test based on the video image edge detection technology is feasible.
出处
《实验技术与管理》
CAS
北大核心
2016年第12期48-52,55,共6页
Experimental Technology and Management
基金
国家自然科学基金项目(51078093
51278137)
广州市科技计划项目(12C42011564)
关键词
连续梁
视频图像
模态分析
continuous beam
video image
modal analysis