摘要
提出一种利用颜色信息进行车道线检测并且能够分辨黄色或白色车道线的新方法。首先,找出图像中与路面颜色差异较大并且具有合理宽度的像素段;然后在RGB颜色空间利用先验信息对像素段的颜色进行辨识;再用辨识后的像素段分别估计出黄色或白色车道线的颜色分割阈值;最后利用获取的阈值对整幅图像进行车道线检测。实验结果表明,该方法能够在复杂背景环境或路面污染等干扰条件下较好地检测出车道线并能辨识出车道线颜色。本方法简单、有效,且具有一定的鲁棒性。
This paper proposes a new lane detection and color identification algorithm using color information. Firstly, the seg- ments with prominent color difference and reasonable width were located. Secondly, the colors of the located segments were iden- tified with color prior in RGB color space. Thirdly, the yellow and white color thresholds were estimated by the identified seg- ments, respectively. Finally, the yellow and white lane markers were segmented with the thresholds. The experiments results showed that the proposed algorithm is able to estimate the thresholds for segmentation and identify the color of lane marker color and applicable in complex environments such as polluted roads or clutter background. The algorithm is simple, effective, and robust.
出处
《计算机与现代化》
2014年第2期86-90,共5页
Computer and Modernization
