Highway guardrail extraction method

The invention discloses a highway guardrail extraction method, which is applied to the technical field of digital image processing, and mainly comprises the following steps of: adaptively carrying out binary image preliminary extraction in an HSV (hue, saturation and value) space through an improved distance formula; removing a large amount of interference semantics through morphological reconstruction and a contour area method; and a Gabor filter and superpixel segmentation are combined to remove similar semantics to obtain a final ROI region image. Compared with the prior art, due to the fact that the background of the road guardrail is complex, the method has the remarkable advantages that robustness is high, the accuracy of detection and recognition is high, the structure of the guardrail image obtained through segmentation is complete, and a good foundation is laid for subsequent detection. 本发明公开了一种公路护栏提取方法，应用于数字图像处理技术领域，主要包括在HSV空间中通过改进的距离公式自适应进行二值化图像初步提取；通过形态学重建和轮廓面积法的方法去除大量的干扰语义；将Gabor滤波器和超像素分割结合起来进行相似