激光光斑边缘提取算法应用研究

Research and application of edge detection algorithm for laser spot

在以激光光斑为被测目标的光电测量系统中,对激光光斑边缘的提取是保证测量系统精度的首要前提。具体介绍了Sobel算子、Krisch算子、Robert算子、Canny算子及LOG算子的基本原理。对实际采集到的激光光斑图像首先进行中值滤波、二值化处理,用于滤除图像中的噪声。然后分别利用上述算子对预处理后的激光光斑进行边缘提取,分析和比较实验结果后得出,在对激光光斑这一特殊的目标进行边缘提取时,LOG算子的提取效果较好。最后,结合实验结果对各算法进行了分析总结。

In the photoelectric measurement system with a laser spot as the research objective,the most important prerequisite to ensure the accuracy of the measuring system is the extraction of laser spot edge.This paper specifically describes the basic principles of Sobel operator,Krisch operator,Robert operator,Canny operator and Laplace of Gaussian operator.In order to filter out the image noise of the captured laser spot images,the Median filtering and Binarization processing are used,then the above operators are used to detect the edge of the pre-processed laser spot images.The analysis and comparison of experimental results show that the Laplace of Gaussian operator is better in this particular edge detection.Finally,the advantages and disadvantages of these methods were demonstrated by discussing the experimented results.