中文说明：在道路图像中， 大部分图像信息对于车道线检测是无用 的， 通过寻找对车道线检测有用的感兴趣区域， 不但可以降低 算法的运算量， 而且能简化车道线的识别。为了提高算法的实时性， 算法还可 以采用动态改变感兴趣区域大小的方法。如果算法识别出的 道路边界可信度较高时， 可进一步缩小感兴趣区域; 而当前一 次算法识别出的道路边界可信度较低时， 则需要对感兴趣区域 进行扩大， 以提高识别的可靠性。
English Description：
In road image, most image information for Lane detection is useless, By finding useful regions of interest for Lane detection, can not only reduceThe computational complexity of the algorithm, and can simplify the identification of lane. In order to improve the timeliness, the algorithm can beTo dynamically change the size of the region of interest. If the algorithm is identifiedRoad borders when credibility is high, you can further narrow the region of interest; And the currentAlgorithms identify the road edge when credibility is low, you need to on region of interestExpanded in order to improve its reliability.