基于位运算的智能车路径检测算法  

Path detection algorithm for smart car navigation based on bitwise operation

作　　者：徐宇宝[1] 林华[1] 罗潇[2] 

机构地区：[1]皖西学院电气与光电工程学院,安徽六安237012 [2]江西理工大学电气工程与自动化学院,江西赣州341000

出　　处：《江西理工大学学报》2017年第1期79-83,共5页Journal of Jiangxi University of Science and Technology

基　　金：皖西学院校级重点研究项目(KJ103762015B05;WXZR201623)

摘　　要：针对智能车视觉寻迹任务,提出一类基于位运算的路径检测算法.在该方法中,首先针对采集到的原始图像容易被噪声污染这一问题,采用基于位运算的高斯滤波来进行噪声抑制;然后,在此基础上分析图像的特点,通过位运算得到图像的位面图来实现图像分割操作;最后对分割后的二值图像进行稀疏扫描和折半查找来获取路径的边界点.对实际采集的路径图像进行路径边界检测的结果表明,所研究的方法可以准确地检测到路径边界,是有效可行的.To perform the task of vision based smart ear navigation, a bitwise operation based on path detection algorithm is proposed in this paper. In the presented method, the bitwise operation based Gaussian filter is adopted to suppress the noise to address the problem of the capturing image which is sensitive to noise. In the next step, bitwise operation is applied to implement image segmenting. Finally, the path boundary is located on the binary segmented image through sparse scanning and binary search techniques. Simulation results on practical path images indicate that the proposed method is both effective and feasible; it can locate the path boundaries accurately.

关 键 词：智能车 位运算 高斯滤波 图像分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]