基于显微镜聚焦的微装配视觉伺服研究

利用显微镜聚焦理论,沿显微镜光轴方向移动物体,不断计算图像的灰度变化之和,可判断出物体沿光轴的坐标,将这一坐标集成在伺服控制方程中,可完成立体视觉跟踪。这样,采用单目视觉系统就可以获得物体的三维坐标,避免了双目立体视觉系统的复杂结构。为了提高系统图像处理速度,利用卡尔曼滤波器对跟踪的特征点进行预测,并用窗口处理技术减小图像处理区域。实验和仿真结果表明,上述方法可完成复杂微装配的视觉跟踪,系统有好的实时性。

Study of Visual Servoing for Microassembly Based on Focus of Microscope

When the object is moved along optical axis of microscope and the variational gray-level of the image is added at the same time,using focus theory,the coordinate of object is obtained along optical axis of microscope,the coordinate of object can be used in servo control equation to finish three dimension vision tracking of microassembly.In this way,when single vision system is used,the three dimension coordinate of object can be obtained and complicated stereo vision system can be avoided.In order to improve the speed of image processing,the Kalman filter is used to predict the next place of tracked point.The results on experimentation and simulation show that complex microassembly task can be finished and the speed of image processing can be improved.