2022
DOI: 10.1155/2022/3825532
|View full text |Cite
|
Sign up to set email alerts
|

Traffic Sign Detection via Improved Sparse R-CNN for Autonomous Vehicles

Abstract: Traffic sign detection is an important component of autonomous vehicles. There is still a mismatch problem between the existing detection algorithm and its practical application in real traffic scenes, which is mainly due to the detection accuracy and data acquisition. To tackle this problem, this study proposed an improved sparse R-CNN that integrates coordinate attention block with ResNeSt and builds a feature pyramid to modify the backbone, which enables the extracted features to focus on important informat… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
32
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
9
1

Relationship

3
7

Authors

Journals

citations
Cited by 64 publications
(32 citation statements)
references
References 39 publications
0
32
0
Order By: Relevance
“…Detection performance in traffic scenes is evaluated using the BCTSDB [39], KITTI [40], and COCO [41] datasets to evaluate the generalization ability. The KITTI dataset contains 7481 training images and 7518 test images, totaling 80,256 labeled objects with three categories (e.g., vehicle, pedestrian, and cyclist).…”
Section: Datasetsmentioning
confidence: 99%
“…Detection performance in traffic scenes is evaluated using the BCTSDB [39], KITTI [40], and COCO [41] datasets to evaluate the generalization ability. The KITTI dataset contains 7481 training images and 7518 test images, totaling 80,256 labeled objects with three categories (e.g., vehicle, pedestrian, and cyclist).…”
Section: Datasetsmentioning
confidence: 99%
“…Detection performance in traffic scenes was evaluated using the BCTSDB [26] and KITTI [27] datasets. BCTSDB is a traffic sign collection that comprises 15,690 images and 25,243 annotations, with labels classified as prohibitory, required, or warning.…”
Section: Datasetmentioning
confidence: 99%
“…The improved deep learning algorithm achieves an average accuracy of 84.22% on established data set. Liang et al [ 21 ] proposed a sparse R-CNN that integrates coordinate attention block with ResNeSt, and better performance is obtained. Ahmed [ 22 ] proposed a new network to enhance traffic sign areas in images of complex environments and used CURE-TSD sets to evaluate the effectiveness of the method, achieving 91.1% accuracy and 70.71% recall rate.…”
Section: Introductionmentioning
confidence: 99%