Lane line detection method for embedded platform

Zhongqiang Du; Linbo Tang; Yuqi Han

doi:10.3788/IRLA20210753

Infrared and Laser Engineering, Volume. 51, Issue 7, 20210753(2022)

Lane line detection method for embedded platform

Zhongqiang Du¹, Linbo Tang¹, and Yuqi Han^2、*

¹Beijing Key Laboratory of Embedded Real-time Information Processing Technology, School of Information and Electronic, Beijing Institute of Technology, Beijing 100081, China

²Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Block diagram of system

Download full size

View in Article

Fig. 2. Lane line detection neural network structure

Fig. 3. [in Chinese]

Fig. 3. Encoder and decoder network

Download full size

View in Article

Fig. 4. Lane line fitting

Download full size

View in Article

Fig. 5. Hisilicon Hi3559 CV100

Fig. 6. DataSets

Fig. 6. [in Chinese]

Fig. 7. Training results of different data sets

Download full size

View in Article

Fig. 8. Lane line detection result

Download full size

View in Article

Table 1. Accuracy of different data sets
View table
View in Article
Table 1. Accuracy of different data sets
Data sets TuSimple CULane Mvirgo TuSimple + Mvirgo
Accuracy 97.5 93.21 98.23 97.8

Table 2. Comparison of different lane line detection network structures

View table

View in Article

Table 2. Comparison of different lane line detection network structures

Method	Accuracy	Time/ms	Data sets
Segnet^[9]	96.78	5.2	tuSimple
Unet^[9]	96.46	4.6	tuSimple
SegNetConvLSTM^[9]	98.52	25	tuSimple
UNetCpnvLSTM^[9]	98.46	20.3	tuSimple
Ours	97.5	1.2	tuSimple
Enet-SAD^[11]	96.64%	13.4	tuSimple
LaneNet^[12]	96.38%	16.6	tuSimple
ERFNet^[10]	95.80%	14.1	tuSimple