摘要
道路边缘的提取是运动检测中最关键的一步。针对目前运动检测过程中更新数据量大、实时性不高的问题,提出了一种新型、快速的道路边缘提取方法。该方法首先利用数学形态学对原图像进行增强,然后通过Canny算子对增强图像进行边缘检测,最后采用Hough变换检测出二值图像中的道路边缘。实验证明,该方法准确性较高,可以减少后期运动检测的运算范围,提高运动检测速率。
Aiming at the problem of large amount of date updates and low real-time,a new method to detect the edge is proposed in this article.This approach processes the input image with mathematical morphology first,and then uses the Canny operator to edge detection for the enhanced image,and deals the binary image with Hough transform to extract the edge of road finally.The experiment results indicate that this method has good accuracy and can simplify the operation of movement detection.It can also improve the detection rate of movement.
出处
《太原科技大学学报》
2010年第3期193-196,共4页
Journal of Taiyuan University of Science and Technology
基金
太原市大学生创新创业专项项目(09122042)
关键词
数学形态学
边缘检测
HOUGH变换
mathematical morphology
the edge detection
Hough transform
