基于模糊图像相关的MEMS动态测量技术

Dynamic Measurement of Parameters of MEMS Based on Correlation of Blurred Image

在基于视觉的微机电系统(MEMS)谐振器动态参数测量中,由于模糊图像合成技术对摄像机采集速率要求不高,且能够得到较高的测量精度,因此在实际测量系统中具有广阔的应用前景。在研究模糊图像相关性的基础上,提出了一种基于模糊图像相关性的MEMS平面运动参数测量方法。该方法通过求取模糊图像一阶微分的自相关函数负峰的间距实现振幅的测量,算法在空域内进行,不受变换域分辨率限制,可以通过插值提高振幅测量精度。实验分析表明,图像中不同区域的相似纹理是产生干扰峰的主要原因,通过在振动方向对图像进行对比度增强,可以减小振动方向的纹理重复度。实验结果表明:该方法是可行的,能精确测量MEMS谐振器的动态参数,并且测量方法简单,只需采集一幅MEMS谐振器在振动稳定状态下的模糊图像即可进行测量,数据处理量小,有利于实现在线测量。

In the measurement of MEMS resonator＇s dynamic parameters, because the fuzzy synthesis technology does not require a high camera image sampling rate to obtain a high measuring accuracy, so in the actual measurement system it finds a wide application. ,Based on the study of the blurred image correlation, this paper proposes a technique of dynamic measurement of MEMS parameters for the exact motion tracing. Through computing the distance between two correlated peaks of the first order differential blurred image of the autocorrelation function, the dynamic amplitude of micro-resonators can be obtained. This algorithm is carried out in the spatial domain, the results will not be affected by the resolution of the transform domain, and the measurement accuracy can be improved by interpolation. And through the experimental analysis, the cause of forming the interference peak is clearly identified, and a solution is found and tested. Experimental results show that in this way, the dynamic parameters of MEMS resonators can be effectively measured, and the method is simple, requiring only to collect a fuzzy image of MEMS resonator in the stable state, a small amount of data, and can be used in the online measurement.