&lt;b&gt;Frontal Obstacle Detection Using Background Subtraction and Frame Registration&lt;/b&gt;

Nakasone, Ryuta; Nagamine, Nozomi; Ukai, Masato; Mukojima, Hiroki; Deguchi, Daisuke; Murase, Hiroshi

doi:10.2219/rtriqr.58.4_298

Cited by 20 publications

(15 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yong [ 9 ] used machine vision to detect obstacles at railway crossings. Ryuta [ 25 ] proposed a method using a monocular camera and image processing for obstacle detection. For the methods that are based on image processing or conventional machine-learning techniques, it is difficult to design a unified method to detect and recognize various objects simultaneously.…”

Section: Related Workmentioning

confidence: 99%

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

Wang

Ping

et al. 2018

Sensors

View full text Add to dashboard Cite

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 modules, i.e., the depthwise-pointwise convolution, the coarse detection module, and the object detection module. Depth-wise-pointwise convolutions are used to improve the detection in real time. The coarse detection module coarsely refine the locations and sizes of prior anchors to provide better initialization for the subsequent module and also reduces search space for the classification, whereas the object detection module aims to regress accurate object locations and predict the class labels for the prior anchors. The experimental results on the railway traffic dataset show that FR-Net achieves 0.8953 mAP with 72.3 FPS performance on a machine with a GeForce GTX1080Ti with the input size of 320 × 320 pixels. The results imply that FR-Net takes a good tradeoff both on effectiveness and real time performance. The proposed method can meet the needs of practical application in shunting mode.

show abstract

Section: Related Workmentioning

confidence: 99%

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

Wang

Ping

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…Sriwardene et al [18] proposed a smart driver assisting system, which could detect obstacles in the rail area. Nakasone et al [19] proposed a railway intrusion detection algorithm using a monocular camera and background subtraction. An image processing method including subtraction, binarization, morphological transformation and segmentation was used to track down moving obstacles at the level crossing [20].…”

Section: Related Work Of Railway Clearance Intrusion Detectionmentioning

confidence: 99%

High-Speed Railway Clearance Intrusion Detection with Improved SSD Network

et al. 2019

View full text Add to dashboard Cite

With the rapid development of high-speed railways, any objects intruding railway clearance will do great threat to railway operations. Accurate and effective intrusion detection is very important. An original Single Shot multibox Detector (SSD) can be used to detect intruding objects except small ones. In this paper, high-level features are deconvolved to low-level and fused with original low-level features to enhance their semantic information. By this way, the mean average precision (mAP) of the improved SSD algorithm is increased. In order to decrease the parameters of the improved SSD network, the L1 norm of convolution kernel is used to prune the network. Under this criterion, both the model size and calculation load are greatly reduced within the permitted precision loss. Experiments show that the mAP of our method on PASCAL VOC public dataset and our railway datasets have increased by 2.52% and 4.74% respectively, when compared to the original SSD. With our method, the elapsed time of each frame is only 31 ms on GeForce GTX1060.

show abstract

“…The following technologies are being developed as they are essential for realizing the train operation control system shown in Fig. 1: 1) algorithms for safely controlling trains based on detailed information on train positions and the conditions of facilities and other components; 2) technologies [8] for collecting detailed transport information in real time to predict train operation; and 3) technologies to automatically calculate train operation plans in real time based on the prediction (Fig. 2).…”

Section: New Train Operation Control System Based On Detailed Positiomentioning

confidence: 99%

“…Using machine learning tech-nology, the system was also tried on humans. In the trial, the system was found to be capable of, depending on lighting conditions, detecting humans as far as 250 m ahead and had a detection rate of 90% or more for humans as far as 180 m ahead [8].…”

Section: Frontal Obstacle Detection Using Integrated Sensorsmentioning

confidence: 99%

<b>Recent Trends and Topics in Research and Development Related to Transportation and Traffic Planning Technology</b>

KAWASAKI¹

2019

QR of RTRI

View full text Add to dashboard Cite

The signalling and transport information division of RTRI aims to contribute to enhancing safety, reliability, and convenience through research and development in signalling systems, communication network technology, transportation planning and traffic management, and condition monitoring technology for railway facilities. This paper outlines the recent trends in research and development aimed at new train operation and control technology for safer and more stable transportation services. This paper further reports on recent development in technology for detecting obstacles on the track, to reduce railway traffic accidents.

show abstract

<b>Frontal Obstacle Detection Using Background Subtraction and Frame Registration</b>

Abstract: Hiroki

Cited by 20 publications

References 3 publications

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

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

High-Speed Railway Clearance Intrusion Detection with Improved SSD Network

<b>Recent Trends and Topics in Research and Development Related to Transportation and Traffic Planning Technology</b>

Contact Info

Product

Resources

About