基于改进Hough变换的车位线识别方法  

Parking slot marking recognition based on improved Hough transform

作　　者：张悦旺[1] ZHANG Yue - wang(School of Telecommunications, Tianjin Vocational Institute, Tianjin 300410, China)

机构地区：[1]天津职业大学电信学院,天津300410

出　　处：《计算机工程与设计》2017年第11期3046-3050,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(61271444);中国职业技术教育学会科研规划重点课题基金项目(201402Z01)

摘　　要：为实现车位线的自动识别检测,提出基于改进Hough变换的车位线识别算法。利用像素局部连接特性剔除噪声像素点影响,基于车位线边缘梯度区间的自适应选择,减少参与Hough变换的像素数,减少背景干扰线段的影响,通过边缘直线间的像素灰度一致关系,进一步确认车位线边缘。实例对比结果表明,对于不同的复杂背景,该算法的车位线边缘检测正确率取得了良好的识别鲁棒性,其运行效率满足系统的实时性要求。Based on improved Hough transform,aparking slot marking recognition method based on improved Hough transform was proposed.The local pixel connection characteristic was used to get rid of noise pixels effect.Based on the adaptive selection of edge gradient interval of parking slot marking,the number of pixels involved in the Hough transform and the influence of background line segments were reduced.The edge lines of the parking slot marking to be identified were identified by judging the gray relation between two edges,and thus the parking slot marking was recognized.The comparative experimental results show that the recognition accuracy of the proposed algorithm is robustness under different kinds of interference background and its operation efficiency meets the real-time requirement of the system.

关 键 词：车位线识别 HOUGH变换 局部像素分布 梯度方向区间 像素灰度一致 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]