Lane line recognition based on improved 2D-gamma function and variable threshold Canny algorithm under complex environment

Houzhong Zhang,Haobin Jiang,Yingfeng Cai,Jiasheng Liang,Xing Xu

doi:10.1177/0020294020952477

Houzhong Zhang, Haobin Jiang + Show 3 more

Open Access

https://doi.org/10.1177/0020294020952477

Copy DOI

Abstract

The visual guidance of AGV (automated guided vehicle) has gradually become one of the most important perception methods. Aiming at the problem that it is difficult to extract lane line accurately when AGV is running in complex working environment (such as uneven illumination, overexposure, lane line is not obvious, etc.), a scheme of lane line recognition under complex environment is proposed. Firstly, the variable scale image correction is carried out for the uneven illumination area in ROI (region of interest), and the threshold of Canny algorithm is adjusted adaptively according to the luminance of ROI region by Fuzzy-Canny algorithm; Secondly, the edge points matching the lane width feature are extracted by the way of aerial view. Finally, the curve fitting method based on RANSAC (Random Sample Consensus) is used to fit a curve with the lowest error rate and then get the lane center curve. The experimental results show that the processing algorithm used in this paper is feasible and effective, has strong robustness and fast computing performance, and can meet the requirements of intelligent AGV in various complex environments.

Highlights

With the in-depth development of ‘‘made in China 2025’’strategy, the whole manufacturing industry is urgently thinking about how to improve the degree of automation and intelligence of enterprises, and quickly keep up with the pace of Intelligent Manufacturing in the world.[1]
In the electromagnetic induction mode, due to the conductor is buried in the ground, once the path is determined, it is difficult to adjust and change; Due to the limited accuracy of gyroscope, the positioning error will accumulate with the increase of the distance; The cost of design, construction and maintenance of magnetic guidance is much higher than other methods; Visual guidance has Jiangsu University, Zhenjiang, Jiangsu, China
Inspired by Retinex algorithm, this paper proposes an enhanced correction algorithm based on Fast Guided filter (FGF) and improved two-dimensional gamma function.[21,22]

Summary

Introduction

With the in-depth development of ‘‘made in China 2025’’strategy, the whole manufacturing industry is urgently thinking about how to improve the degree of automation and intelligence of enterprises, and quickly keep up with the pace of Intelligent Manufacturing in the world.[1]. Edge point filtering and recognition algorithm based on lane width feature Through the introduction of adjusting the threshold value in Fuzzy-Canny algorithm, we can see that even though enough edge points on the lane line are identified, there are still two problems: one is how to identify which scatter points belong to the lane line and which scatter points are useless noise points; the other is how to fit the edge points belonging to the lane line into a straight line or curve To solve these two problems, the ROI area in the original image collected by the camera is transformed into an aerial view, the edge points of the lane line are extracted by a feature point screening method based on the lane line width, and the edge points are fitted by the Bezier curve fitting algorithm based on RANSAC algorithm to get the identified lane line.

M1M2664

Findings

Conclusion