微操作中三维运动定位的目标识别误差矫正

Error Correction for 3-D Motion Positioning of Object Recognition in Micromanipulation

根据形状差异和广义图像误差对微操作定位精度的影响,给出了形状差异的矫正算法。 基于光照度影响图像亮度的事实,导出了矫正广义图像误差的识别调节系数。使用精度±2m操作手控制运动,视觉模型的定位精度在5%左右。通过误差矫正,精度可以继续得到提高,提高的程度与识别调节系数的大小有关,而识别调节比达到初始调节比的60%-80%为宜。

A correction algorithm for shape difference is given according to the influence of shape difference and generalized image error on positioning accuracy of micromanipulation. A recognition and adjustment coefficient for correcting generalized image error is derived on the basis of light illuminance affecting brightness. Using a manipulator with an accuracy of ±2 μm to control movement, the positioning accuracy of the vision model is about 5%. The accuracy can be continuously improved through error correction. The improvement level relates to the size of recognition and adjustment coefficient, the recognition-and-adjustment ratio should be equivalent to 60%-80% that of the initial adjustment ratio.