为了实时、准确地提取作物行基准线,提出了一种将边缘检测和扫描滤波(Boundary detection and scan-filter,BDSF)相结合的基准线提取方法。首先对RGB颜色空间采用G-R颜色特征因子进行图像灰度化,再采用最大类间方差法(OSTU)对灰度图像...为了实时、准确地提取作物行基准线,提出了一种将边缘检测和扫描滤波(Boundary detection and scan-filter,BDSF)相结合的基准线提取方法。首先对RGB颜色空间采用G-R颜色特征因子进行图像灰度化,再采用最大类间方差法(OSTU)对灰度图像进行分割,得到二值化图像,获取较好的作物信息。然后分别对图像的底端和顶端部分进行垂直投影,获取作物行的位置,形成一个包含作物行直线的条形框;在这个条形框内,再用等面积的小条形框对图像进行扫描并统计有效点的个数。最后根据扫描的结果来提取导航线。试验结果表明,对比Hough算法和最小二乘法(Least square method,LSM),BDSF算法处理一幅分辨率为640像素×480像素的图像,平均耗时为67 ms,与LSM算法耗时相当,精度接近Hough算法;并且在杂草和株数稀缺情况下具有良好的适应性,能够快速准确地提取作物行基准线。展开更多
In order to obtain a compact and exact representation of 2D range scans,UKF(unscented Kalman filter) and CDKF(central difference Kalman filter) were proposed for extracting the breakpoint of the laser data. Line extra...In order to obtain a compact and exact representation of 2D range scans,UKF(unscented Kalman filter) and CDKF(central difference Kalman filter) were proposed for extracting the breakpoint of the laser data. Line extraction was performed in every continuous breakpoint region by detecting the optimal angle and the optimal distance in polar coordinates,and every breakpoint area was constructed with two points. As a proof to the method,an experiment was performed by a mobile robot equipped with one SICK laser rangefinder,and the results of UKF/CDKF in breakpoint detection and line extraction were compared with those of the EKF(extended Kalman filter) . The results show that the exact geometry of the raw laser data of the environments can be obtained by segmented raw measurements(combining the proposed breakpoint detection approach with the line extraction method) ,and method UKF is the best one compared with CDKF and EKF.展开更多
基金Project(2003AA1Z2130)supported by the National High-Tech Research and Development Program of ChinaProject(2005C11001-02)supported by the Science and Technology Project of Zhejiang Province,China
文摘In order to obtain a compact and exact representation of 2D range scans,UKF(unscented Kalman filter) and CDKF(central difference Kalman filter) were proposed for extracting the breakpoint of the laser data. Line extraction was performed in every continuous breakpoint region by detecting the optimal angle and the optimal distance in polar coordinates,and every breakpoint area was constructed with two points. As a proof to the method,an experiment was performed by a mobile robot equipped with one SICK laser rangefinder,and the results of UKF/CDKF in breakpoint detection and line extraction were compared with those of the EKF(extended Kalman filter) . The results show that the exact geometry of the raw laser data of the environments can be obtained by segmented raw measurements(combining the proposed breakpoint detection approach with the line extraction method) ,and method UKF is the best one compared with CDKF and EKF.
