Journals
Advanced Photonics
Photonics Insights
Advanced Photonics Nexus
Photonics Research
Advanced Imaging
View All Journals
Chinese Optics Letters
High Power Laser Science and Engineering
Articles
Optics
Physics
Geography
View All Subjects
Conferences
CIOP
HPLSE
AP
View All Events
News
About CLP
Search by keywords or author
Login
Registration
Login in
Registration
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
metasurface
laser
polarization
nir
lithium niobate
optical coherence tomography
Journals >
Laser & Optoelectronics Progress >
Volume 57 >
Issue 18 >
Page 181015 > Article
Laser & Optoelectronics Progress
Vol. 57, Issue 18, 181015 (2020)
Traffic Sign Detection Based on Improved Faster R-CNN Model
Yi Zhang, Zhiyuan Gong
*
, and Wenwen Wei
Author Affiliations
School of Communication and Information Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
show less
DOI:
10.3788/LOP57.181015
Cite this Article
Set citation alerts
Yi Zhang, Zhiyuan Gong, Wenwen Wei. Traffic Sign Detection Based on Improved Faster R-CNN Model[J]. Laser & Optoelectronics Progress, 2020, 57(18): 181015
Copy Citation Text
EndNote(RIS)
BibTex
Plain Text
show less
Fig. 1.
Framework of Faster R-CNN
Download full size
Fig. 2.
Block of ResNeXt
Download full size
Fig. 3.
Block of improved basic network
Download full size
Fig. 4.
Basic network structure of multi-dimensional feature fusion
Download full size
Fig. 5.
RPN structure
Download full size
Fig. 6.
Specific parameters of anchor frame type
Download full size
Fig. 7.
Samples of TT100K data set. (a) image 1; (b) image 2; (c) image 3; (d) image 4
Download full size
Fig. 8.
Training set attributes. (a) Number of samples in training set; (b) images corresponding to some traffic sign numbers
Download full size
Fig. 9.
Loss function curves
Download full size
Fig. 10.
P-R curves. (a) Improved basic network; (b) improved basic network+multi-scale feature fusion; (c) improved basic network+multi-scale feature fusion+improved anchor generation
Download full size
Fig. 11.
Example of detection effect of proposed algorithm. (a) Complex background detection; (b) detection of small targets
Download full size
Stage
Output size
Network structure
Conv1
1024×1024
Convolution kernel size: 7×7, 64, stride: 2;
max pooling size: 3×3, 64, stride: 2
Conv2
512×512
1
×
1,128
1
×
1,3
×
3,5
×
5,7
×
7,128,1
,
C
=
4
×
8
1
×
1,256
×
3
Conv3
256×256
1
×
1,256
1
×
1,3
×
3,5
×
5,7
×
7,256,1
,
C
=
4
×
8
1
×
1,512
×
3
Conv4
128×128
1
×
1,512
1
×
1,3
×
3,5
×
5,7
×
7,512,1
,
C
=
4
×
8
1
×
1,1024
×
3
Conv5
64×64
1
×
1,1024
1
×
1,3
×
3,5
×
5,7
×
7,1024,1
,
C
=
4
×
8
1
×
1,2048
×
3
Table 1.
Basic network parameters
Name
Configuration
CPU
Intel Core i7 8700K
GPU
NVIDIA RTX 2080Ti
RAM/G
32
Hard disk /TB
2
Table 2.
Hardware configuration parameters
Algorithm
AP /%
Time /s
Traditional Faster R-CNN
80.28
0.33
Ref.[19]
81.13
0.31
Ref.[8]
91.67
5.81
Proposed algorithm
90.83
2.87
Table 3.
Performance comparison of detection algorithms
Abstract
Get PDF(in Chinese)
Figures&Tables (14)
Equations (0)
References (20)
Cited By (2)
Get Citation
Copy Citation Text
Yi Zhang, Zhiyuan Gong, Wenwen Wei. Traffic Sign Detection Based on Improved Faster R-CNN Model[J]. Laser & Optoelectronics Progress, 2020, 57(18): 181015
Download Citation
EndNote(RIS)
BibTex
Plain Text
Set citation alerts for the article
Tools
Share
Set citation alerts for the article
Save the article for my favorites
Paper Information
Category: Image Processing
Received: Dec. 6, 2019
Accepted: Feb. 24, 2020
Published Online: Sep. 2, 2020
The Author Email: Gong Zhiyuan (gongzy0728@163.com)
DOI:
10.3788/LOP57.181015
Recommended Topics
laser devices and laser physics
Lasers and Laser Optics
Laser physics
laser manufacturing
Instrumentation, Measurement and Metrology
Set citation alerts for the article
Please enter your email address
Cancel
Confirm