摘要
针对大尺寸铝合金铸件的视觉测量,研究了图像拼接方法。首先利用经过标定的摄像机采集多幅具有重叠区域的序列图像,然后使用Harris和RANSAC算法提取精确特征点,并进行相邻图像重叠区域的特征点匹配,生成了完整的工件图像,最后,通过实验验证了该方法的可行性。
In this article,a method of image splicing was studied in allusion to the visual measurement of large-sized aluminum alloy castings.The method firstly captured several sequence images of overlapping regions via a calibrated camera.Then it detected the feature points by using Harris and RANSAC algorithm.Thirdly,the feature points of the adjacent image overlaps were matched so that the complete image of the artifact was generated.Finally,the feasibility of this method was verified by experiment.
出处
《机械工程与自动化》
2018年第1期14-15,18,共3页
Mechanical Engineering & Automation
基金
湖北省自然科学基金资助项目(2014CFB637)
"机电汽车"湖北省优势特色学科群开放基金资助项目
关键词
视觉测量
特征匹配
图像拼接
大尺寸铸件
vision measurement
feature match
image mosaic
large-scale casting
