基于显微图像散焦特征的微操作机器人深度信息提取

MICROMANIPULATION ROBOT DEPTH EXTRACTION BASED ON MICRO IMAGE DEFOCUS FEATURE

提出了一种单目显微视觉下微操作机器人深度信息的快速提取方法 ,通过计算微操作空间中 X- Y平面的散焦图像特征 ,表征图像清晰程度 ,从而获得机械手在 Z方向的深度信息 .实验特征曲线证明方法简洁快速 ,有效的反映了微操作深度信息的变化和机械手在纵向的运动特性 .

This paper presents a depth extraction method for micromanipulation robot based on monocular micro vision. With the depth description in micro manipulating space, one kind of image features has been extracted from 2-D image plane to evaluate defocusing. It is demonstrated by experimental results that the method is available and the determined defocus feature is consistent with the robot movement property in Z direction.