基于多尺度分析的激光光条中心点坐标提取方法

A Multi-Scale Analysis Based Method for Extracting Coordinates of Laser Light Stripe Centers

在线结构光测量系统中,激光光条图像中心准确提取是影响整个测量系统精度的关键因素之一。针对高反光等复杂环境下光学测量中激光光条图像中心提取问题,提出了一种基于多尺度分析的激光光条图像中心高精度定位方法。该方法利用骨骼化图像处理方法确定激光光条图像中心的初值,根据每一个光条中心初值处的光条宽度确定该位置处对应的高斯核均方差σ的初值。对图像进行多尺度卷积,确定最优的σ值,并求得光条亚像素级中心点,最后完成光条链接。结果表明该算法抗噪声能力强,可实现光条宽度变化较大的激光光条图像中心的高精度提取。

In a line structured light measurement system, the accurate extraction of the light stripe centers is one of the key factors affecting the accuracy of the entire system. For the problem of light stripe center extraction in optical measurements in complex environment with high reflection, a multi-scale analysis based method that realizes high- precision extraction of light stripe centers is proposed. The skeleton algorithm is applied to determine the initial centers of the laser stripe, and the mean square deviation a of the Gaussian kernel is determined based on the width of the light stripe at each initial center. Multi-scale image convolution is carried out to determine the optimal σ, in order to further improve the accuracy of extracting light stripe center. Interlinkage of the light stripe centers is finished. The results show that the method is of high noise immunity, and can achieve high-precision results when the width of the light stripe changes sharply.