2018
DOI: 10.3390/s18061916
|View full text |Cite|
|
Sign up to set email alerts
|

Automatic Railway Traffic Object Detection System Using Feature Fusion Refine Neural Network under Shunting Mode

Abstract: Many accidents happen under shunting mode when the speed of a train is below 45 km/h. In this mode, train attendants observe the railway condition ahead using the traditional manual method and tell the observation results to the driver in order to avoid danger. To address this problem, an automatic object detection system based on convolutional neural network (CNN) is proposed to detect objects ahead in shunting mode, which is called Feature Fusion Refine neural network (FR-Net). It consists of three connected… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
27
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
7
3

Relationship

1
9

Authors

Journals

citations
Cited by 51 publications
(31 citation statements)
references
References 37 publications
0
27
0
Order By: Relevance
“…Both are computed as the sum of squared errors. In testing process, the non-maximum suppression (NMS) method [ 31 ] is applied to get rid of redundant bounding boxes.…”
Section: Methodsmentioning
confidence: 99%
“…Both are computed as the sum of squared errors. In testing process, the non-maximum suppression (NMS) method [ 31 ] is applied to get rid of redundant bounding boxes.…”
Section: Methodsmentioning
confidence: 99%
“…CNNs have also been utilised to perform railcar safety inspection [154], determine the area of the rails ahead [155] detecting objects ahead [156], detect multiple catenary systems and support components [157]- [159], tracking joints [160] and detecting track defects [161], [162].…”
Section: Related Work In Railway Systemsmentioning
confidence: 99%
“…Hu et al [35] combined traditional CNN with a spatial pyramid strategy to recognize the vehicle color under various conditions, but the accuracy was poor at night. Ye et al [36] proposed a CNN-based railway object detection network consisting of three connected modules, i.e., a depthwise-pointwise convolution module, coarse detection module and object detection module. Railway experiments showed that the network could distinguish straight, railway left, right, pedestrian, bullet train, and safety helmets.…”
Section: Object Detection With Cnnsmentioning
confidence: 99%