基于机器视觉边缘检测的园林喷药机器人导航线提取

Extraction of navigation line for garden spraying robot based on machine vision edge detection

导航路径的精确拟合是园林机器人自动化导航的关键,针对现有园林喷洒机器人仍是人工操作为主的现象,提出一种基于视觉边缘检测的导航路径拟合算法,用于指导园林喷药机器人的自动化导航。首先利用“化曲为直”的思想,截取拍摄图像的最后200像素行作为感兴趣区域;其次提出一种针对园林道路的灰度化因子,对图像进行灰度化处理;然后对图像进行行扫描,根据噪声和道路在几何学上的差异,设定宽度阈值T,快速检测出道路边缘;根据检测出的边缘点,计算道路边缘点的均值拟合出边界线;最后,根据道路顶点坐标,计算出梯形道路的中轴线方程提取出导航线,为园林机器人导航提供参数。试验结果表明该算法处理一幅图像平均耗时53 ms,误差在0.6°内,可见本文所提算法可以满足实时性要求,且算法鲁棒性好、准确性高,为园林喷药机器人的自动化导航提供理论依据。

The accurate fitting of navigation path is the key to the automatic navigation of garden robot.Aiming at the phenomenon that the existing garden spraying robot is still dominated by manual operation,this paper proposes a navigation path fitting algorithm based on visual edge detection to guide the automatic navigation of garden spraying robot.Firstly,the idea of“turning curve into straight”is used to intercept the last 200 pixel line of the image as the region of interest;Secondly,a graying factor for garden road is proposed to gray the image;Then the image is scanned,and the width threshold T is set according to the difference between noise and road in several aspects,so as to detect the road edgequickly;According to the detected edge points,the mean value of road edge points is calculated to fit the boundary line.Finally,according to the coordinates of road vertex,the central axis equation of trapezoid road is calculated,and the guide line is extracted,which provides parameters for the automatic navigation of garden robot.The experimental results show that the average time of the algorithm is 53 ms,and the error is within 0.6°.It can be seen that the proposed algorithm can meet the requirements of real-time,and the algorithm has high robustness and accuracy,which provides a theoretical basis for the automatic navigation of garden spraying robot.