【摘要】 为满足在高精细导航电子地图中进行车道线绘制的需求,本文提出一种基于概率霍夫变换的车道线识别算法。算法在图像预处理部分使用一种基于转换颜色空间的方法提取车道线像素,之后利用自适应阈值的Canny边缘检测算法和概率霍夫变换算法实现车道线识别。试验结果表明,本文算法的运算速度维持在2秒/帧;识别正确率与Matlab经典Hough算法相比提升9%左右,漏检率也有所降低。同时拍摄质量较差图像的针对性试验结果,也证实了本文算法能有效降低天气和光照等因素对车道线识别造成的影响。更多还原

【Abstract】 In order to meet the requirements of lane line drawing in high-precision navigation electronic map,a lane recognition algorithm based on probabilistic Hough transform is proposed. In the image preprocessing part,the proposed algorithm extracts lane pixels based on the method of transforming color space. Then the lane recognition is realized by using Canny edge detection algorithm with adaptive threshold and probability Hough transform algorithm. Experiments show that the computational speed of the algorithm is maintained at 2 seconds per frame. In terms of accuracy,the recognition accuracy rate is improved by about 9% compared with the Matlab classic Hough algorithm,and the missed detection rate is also reduced. It is also proved that the proposed algorithm can effectively reduce the influence of weather and illumination on lane recognition.更多还原