Artificial intelligence-aided railroad trespassing detection and data analytics: Methodology and a case study

Zhang, Zhipeng; Zaman, Asim; Xu, Jinxuan; Liu, Xiang

doi:10.1016/j.aap.2022.106594

Cited by 25 publications

(13 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the number of crashes and fatalities at grade crossings is significant, it is the result of a series of precursory risky behaviors. Research performed by Zaman et al ( 7 , 8 ) and Zhang et al ( 9 , 10 ) from 2018 to 2022 has demonstrated that there are many more grade crossing violation incidents which do not result in accidents. However, each of these events has the potential to result in a fatality.…”

Section: Literature Reviewmentioning

confidence: 99%

Artificial Intelligence-Aided Grade Crossing Safety Violation Detection Methodology and a Case Study in New Jersey

Zaman

Huang

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

Fatalities at grade crossings accounted for an average of 33% of all railroad industry fatalities occurring in the past 10 years. As road traffic increases and high-speed rail deployments become more common in the United States, the number of fatalities is expected to remain a concern. Railroads have tackled this challenge through a combination of engineering, education, and enforcement campaigns. One of these efforts has been the increased deployment of security cameras throughout railroad networks. These video sources allow for the collection of big data to better understand grade crossing violation behaviors. However, monitoring these video feeds and extracting useful information requires prohibitive amounts of manual labor. This research utilizes state-of-the-art vision-based artificial intelligence (AI) techniques to record, recognize, and understand railroad video data in real time. This system’s understanding of active grade crossing violations helps to develop precise long-term grade crossing violation prevention strategies. This study explains how this AI-aided algorithm is used to monitor 1 year’s worth of violations at an active grade crossing in New Jersey and provides an overview of the observed trends. These data can be used to develop better engineering enforcement and education strategies for the mitigation of active grade crossing violations.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Artificial Intelligence-Aided Grade Crossing Safety Violation Detection Methodology and a Case Study in New Jersey

Zaman

Huang

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly, Zhang et al. [36] developed a robust deep learning framework for the automatic detection of trespassing events.…”

Section: Related Workmentioning

confidence: 99%

“…The accurate detection of moving objects was achieved via the fusion of a detector based on GMM and a classifier of Alexnet, while the traffic velocity was estimated by Lucas-Kanade optical flow. Similarly, Zhang et al [36] developed a robust deep learning framework for the automatic detection of trespassing events.…”

Section: Related Workmentioning

confidence: 99%

Whether and how congested is a road? Indices, updating strategy and a vision‐based detection framework

Gao

You

Wang

et al. 2023

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Traffic congestion detection method based on surveillance video is gradually widely used in intelligent transportation systems (ITS). Due to complex challenges such as weather change, vehicle occlusion, camera jitter, camera installation location, and so on, current methods are difficult to balance in real time and accuracy. Here, a new real‐time and robust traffic congestion detection framework and a vision‐based multi‐dimensional congestion detection model are proposed. Firstly, the framework introduces an object detector based on the lightweight convolutional neural network (CNN) and an efficient multi‐object IoU‐like tracker to obtain traffic dynamic information in real time. Then, traffic density, traffic velocity, and duration of instantaneous congestion are defined and a multi‐dimensional congestion detection model is established. Furthermore, an adaptive updating strategy of dynamic parameters is investigated. Finally, in multiple groups of comparative experiments, the framework is verified to be applicable to a variety of lightweight CNN detectors and IoU‐like trackers. The precision and recall of the multi‐dimensional congestion model can reach 95.1% and 92.1% respectively, with 43FPS. The comparative experimental results show that the proposed method is real‐time, more robust and accurate, and can be employed for online traffic congestion detection based on surveillance video.

show abstract

“…Wang introduced a novel neural network based on the SSD framework for railway intrusion detection ( 19 ). Zhang et al developed a YOLO-based framework for the automatic detection of railroad trespassing events ( 20 ). However, as mentioned in the introductory section, most of these studies used CNN-based object detectors that can only detect ordinary objects such as cars and pedestrians and have a problem detecting unusual outliers.…”

Section: Related Workmentioning

confidence: 99%

Railroad Crossing Surveillance and Foreground Extraction Network: Weakly Supervised Artificial-Intelligence Approach

Tang

Wang

Qian

2023

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

According to the report to Congress National Strategy to Prevent Trespassing on Railroad Property issued by the Federal Railroad Administration, trespassing is currently the number one cause of all railroad-related deaths. The number of fatalities resulting from trespassing, including both illegally entering and remaining in the railroad right-of-way, is even higher than the number of deaths caused by collisions between vehicles and trains. Accurately and effectively detecting objects at rail crossings is critical for improving railway safety. The convolutional neural network (CNN) is one of the most potent tools in image processing and video surveillance. However, the current CNNs under supervised learning rely on a substantial amount of manually labeled images, which is labor-intensive and time-consuming. This study proposes a weakly supervised learning technique for training foreground-segmentation networks and develops a tailored CNN network, RC-SAFE (railroad crossing surveillance and foreground extraction), to accurately and effectively detect and track both still and moving objects in the railroad crossing area. The weakly supervised learning significantly reduces the load for image labeling. The experimental results on the public CDnet 2014 dataset show that the weakly supervised network outperforms the supervised network and some other state-of-the-art foreground-segmentation methods. Railroad crossing monitoring videos are collected at different railroad crossings to validate the proposed RC-SAFE network. The results confirm the robustness of the proposed RC-SAFE network, and that RC-SAFE can detect any object that does not belong to the railroad crossing, which outperforms object-detection-based models, such as Mask RCNN (a region-based CNN).

show abstract

Artificial intelligence-aided railroad trespassing detection and data analytics: Methodology and a case study

Cited by 25 publications

References 19 publications

Artificial Intelligence-Aided Grade Crossing Safety Violation Detection Methodology and a Case Study in New Jersey

Artificial Intelligence-Aided Grade Crossing Safety Violation Detection Methodology and a Case Study in New Jersey

Whether and how congested is a road? Indices, updating strategy and a vision‐based detection framework

Railroad Crossing Surveillance and Foreground Extraction Network: Weakly Supervised Artificial-Intelligence Approach

Contact Info

Product

Resources

About