利用普通摄像头实现振动的实时测量

Real Time Measurement of Vibration Based on USB Camera

为实现对低频小振幅振动的实时测量,以USB摄像头为图像传感器,以常规计算机为上位机,设计了一套性价比较高的基于数字图像的检测装置。介绍了检测系统的原理和结构,以及主要的图像处理方法。根据系统的特点,本系统应用基于颜色的图像分割方法,大大提升了系统的适应性和可靠性,并利用捕捉振动物体部分区域的方法提高系统的测量精度,同时将卡尔曼滤波方法用于预测振动物体的位置,大大降低了系统处理的图像数据量,提高了系统的实时性,进而提高了系统的测量范围。二维低频振动测量实验证明了本系统性能可靠。

In order to measure low-frequency low-amplitude vibration in real-time, this paper designs a high performance-cost ratio set of image measurement system, which uses USB camera as image sensor and PC as system processor. The principle , structure and methods of image processing of system are presented. According to the special task of this system, the image segmentation method based on color threshold is adopted, which greatly improves its adaptability and reliability, and the accuracy is increased due to capturing one portion of vibrating object image. Using Kalman filter to predict the position of object dramatically brings down the amount of image data and raises the real time processing performance of system. Moreover, the reliability of system is varified by experiments on 2-D vibration measurement.