Fast lane recognition based on morphological multi-struc-ture element model

• 来源期刊: 光电子快报（英文版）
• 年，卷（期）: 2009,（4）
• 页码范围: 304-308
• 页数: 5页
• 语种: 英文
• DOI: 10.1007/s11801-009-8183-y

摘要

This paper proposes a lane detection algorithm based on multi-structure element model of morphological. The innovative point of the algorithm lies in the facts that the flexible structure has the multi-structure elements that lane model features have, and that the algorithm adopts the morphological filtering principle to extract the pixels in the image, which is similar to the lane model. In the algorithm, the interested area is extracted by a model of trapezium from original image, which is detected by the operator of Canny, and the lanes are extracted by the structure elements, which have similar characteristics to that of lane model. Several lines are detected by Hough transformation, then the traffic lanes are reconstructed. Experi-ments show that this algorithm is simple and robust, and can efficiently detect the lane mask accurately and quickly.