复杂工况下视觉AGV导航标识带中心线的提取研究

Study on Center Line Extraction for AGV’s Guide Ribbon Under Complex Conditions

摘要:

针对光照不均匀、导航标识带破损、少量杂物干扰的复杂工况条件下,对视觉AGV采集到的路面导航标识带图片进行特征提取,采用灰度化、中值滤波进行了图像预处理,研究了Otsu法和迭代法等动态阈值分割算法以及形态学在图像分割中的应用,并对预处理图像作了分割对比实验;然后对分割后的图像进行边缘提取,分析了基于最小二乘法和Hough变换算法的直线拟合原原理,提出了Hough算法下基于边缘线的中心线拟合算法,并作了直线拟合对比试验;实验表明,复杂工况下,采用基于形态学和Otsu算法相结合的方法对图像进行分割,得到的二值图像边界更完整,效果更好,基于Hough变换算法较最小二乘法能更精确有效地提取出导航标识带中心线及其方程。

Abstract:

Aiming at the complex condition of non-uniform lighting,damaged ribbon and a few clutter interference , the guided ribbon pictures of AGV based on vision has been collected to extract features, conducting the image pretreatment by using the methods of Graying and Median filter,study on the dynamic threshold segmentation methods include Otsu algorithm and Iterative method and the application of mathematical morphology in image processing, and make image threshold segmentation experiments respectively to analysis.Then extract the boundary of the segmented image ,analysis on the least-square and Hough transform methods ,a algorithm of ribbon's center line fitting based on the Hough transform is proposed , and using the methods respectively to obtain ribbon's center line and its mathematical Equation .Experimental results show that under the complex conditions ,using the method based on Otsu algorithm and mathematical morphology to segment image can get better effect on Boundary integrality of binary image ,and using the algorithm based on Hough transform to extract ribbon's center line and its mathematical Equation can be more accurate and effective.