亚像素条纹边缘提取技术在型面测量上的应用

三维型面非接触测量在工业领域有着广泛的应用前景.双目视觉是一种有效获取三维轮廓信息的方法,其中对大密度的黑白相间条纹进行精确定位和提取,是基于投影光栅图像的光学三维型面测量的关键技术和难点.利用空间编码技术将被测空间划分为若干区域,每个区域对应一个二进制码.条纹落在区域上与唯一的二进制码相对应,就可以对条纹精确定位.利用边缘提取算法获得像素级条纹后,再利用亚像素技术可以进一步对条纹进行高精度提取.介绍了基于灰度矩的和基于多项式拟合的亚像素边缘检测算法.将空间编码和亚像素条纹提取技术相结合,对光栅条纹进行亚像素级定位和提取的技术,与像素级的定位技术相比,能提高图像的定位精度,有效地改善测量系统的性能. 

The technique of sub-pixel stripes distinguish on the surface measurement

Non-contact measurement of 3-D surface has been extensively applied in industry.The stereo vision is an effective way for 3-D profile measurement.Distinguishing and detecting the high density white-black stripes is one of the key techniques in the 3D-measurement based on the image of grating.Using spatial code technology,the measurement space was divided to several parts,and each part corresponded with a unique binary code.Falling on the space part,there was a unique binary code corresponded with the stripes,thus the stripes could be distinguished accurately.After getting pixel stripes with edge detection algorithm,the high precision strips could be detected by sub-pixel technology.The sub-pixel edge detection algorithm based on gray moment and polynomial fitting was presented.Combining spatial code and sub-pixel technique,it could easily distinguish and detect the sub-pixel stripes.This method can obtain a higher precision comparing with pixel stripes detection and improve the accuracy of measurement system.