Method, device, readable storage medium and electronic equipment for lane line recognition

Abstract

The embodiment of the invention discloses a lane line recognition method, device, readable storage medium and electronic equipment. In the embodiment of the present invention, by clustering the pixels in the lane line area according to the color, the color type corresponding to different cluster data is determined according to the center point of the cluster data, and the color type in the lane line area is determined according to the color type corresponding to the cluster data. Included color type, if a lane line area includes yellow and gray, or white and gray; then determine the color type of the lane line in the lane line area, if the lane line area includes yellow and gray, then the color type of the lane line is yellow, and the color type of the background area is gray; and the lane line area is binarized, and the type of the lane line is determined according to the proportion, for example, the lane line is a solid line or a dashed line; through the above method, the accurate Determines the color type of lane markings in the exit lane marking area and the line type of lane markings.

Description

technical field [0001] The present invention relates to the field of unmanned driving, in particular to a lane line recognition method, device, readable storage medium and electronic equipment. Background technique [0002] With the rise of unmanned driving technology, traffic scene detection is becoming more and more important. Traffic scene detection needs to meet accurate requirements. Traffic scene detection includes the identification of lane line types, and the identification of lane line types is an important aspect of advanced driver assistance systems (Advanced Driver Assistance Systems). At the core of the Driver Assistance System (ADAS), the lane line types are divided into white lines and yellow lines from the color, and solid lines and dashed lines from the line type. [0003] In the prior art, the yellow line and the white line in the image are distinguished by selecting a color range in a fixed color space, and the image is segmented by a neural network model ...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the yellow line and the white line in the image are distinguished by selecting a color range in a fixed color space, and the image is segmented by a neural network model to realize the identification of a dotted line and a solid line; due to different weather, or The images acquired by different cameras have color differences, so it is impossible to accurately distinguish the yellow, white and gray pixels in the image with a fixed color range. For example, the actual color is gray, but due to the influence of the weather, the actual The gray part of the pixel in the image falls within the white color range, and it is recognized as white, resulting in a color difference, which makes it impossible to distinguish the yellow line from the white line; and the image is segmented by color. Due to the inaccurate color judgment, the image segmentation is also inaccurate, and the determined dotted and solid lines are also inaccurate

Images