Road Scene Segmentation Based on Deep Learning

Ke Zheng,Hasan Abdullah Hasan Naji

doi:10.1109/access.2020.3009782

Ke Zheng, Hasan Abdullah Hasan Naji

Open Access

https://doi.org/10.1109/access.2020.3009782

Copy DOI

Abstract

Recently, autonomous driving becomes a hot topic in research and industry area. Autonomous driving technology needs to perceive the semantic information of road scenes in the all-day and open environment. In this article, the semantic recognition of traffic scenes is studied using a deep learning network model, and a semantic representation model of road scenes is established. Besides, a semantic recognition algorithm of road scenes based on image data is proposed. Finally, a self-built data set is used to train the proposed model, and verified in the field of the test vehicle. It is found that the proposed method can quickly capture the perceptual road scene and over-performs better than traditional methods and demonstrated great potential to be used in road scene applications.

Highlights

In recent years, artificial intelligence technologies represented by deep learning [1] have promoted the development of autonomous driving, which has become a key technology in the development planning of the first generation of artificial intelligence in various countries
Sensing the state of the road is crucial in self-driving technology and at present, many researchers accurately detect and identify drivable area [3], lane lines [4], vehicles and pedestrians [5], traffic lights [6], traffic signs [7] and other environmental information by using the data captured by vehicle sensors, providing basis for decision-making and control of automatic driving system
This paper proposes a deep learning neural network ROADNET for road scenes and establishes a semantic representation model of road scenes, to realize the semantics of road scenes automatically perceived by vehicles

Summary

Introduction

Artificial intelligence technologies represented by deep learning [1] have promoted the development of autonomous driving, which has become a key technology in the development planning of the first generation of artificial intelligence in various countries. Autonomous driving [2] is committed to provide humanized intelligent control to vehicles, and has great research value. It has received extensive attention from academia and industry area. Sensing the state of the road is crucial in self-driving technology and at present, many researchers accurately detect and identify drivable area [3], lane lines [4], vehicles and pedestrians [5], traffic lights [6], traffic signs [7] and other environmental information by using the data captured by vehicle sensors, providing basis for decision-making and control of automatic driving system. Researches on the road scene of automatic perception of vehicles are rare at present

Methods

Findings

Conclusion