一种适用于特征稀少零件的图像精确拼接方法

An accurate image mosaic method for part with rare feature information

针对特征稀少零件的图像精确拼接难题,提出了一种基于相位相关法和闭环运动控制的图像精确拼接方法,以充分发挥软硬件的综合优势。该方法以具有足够特征信息的零件为对象,获取成像系统分别沿X轴和Y轴运动时的零件等距序列图像并预处理,再利用相位相关法求解图像配准参数。在闭环运动控制系统的良好重复定位精度支持下,将上述图像配准参数视为系统配准参数,以进行特征稀少零件的图像拼接。典型零件的图像拼接实验表明,该拼接方法用于特征稀少零件的图像拼接具有无像素级拼接错位和拼接速度快等优点。

Focused on the problem that part image with rare feature information is hard to mosaic,an accurate image mosaic method based on phase correlation algorithm and closed loop motion control is proposed to make full use of the comprehensive advantages of the software and hardware.With a part with enough feature information as an object,this method captures serial images of the part when imaging system moves along X axis and Y axis respectively and does pretreatment,and then solves registration parameters using phase correlation algorithm on two adjacency images.With the support of good repeatability of closed loop motion control system for vision measuring machine,the solved image registration parameter can be used as system registration parameter to mosaic image of the part with rare feature information based on system registration parameter directly.Image mosaic experiment of typical part shows the proposed image mosaic method has advantages that the image mosaic for part with rare feature information not only has no misregistration at pixel level but also is rapid in mosaicing.