On the safety of vulnerable road users by cyclist detection and tracking

Garcia-Venegas, Marichelo; Mercado-Ravell, Diego Alberto; Pinedo-Sanchez, Luis A.; Carballo-Monsivais, Carlos A.

doi:10.1007/s00138-021-01231-4

Cited by 10 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This research works under the assumption that the cyclist has been detected from the LiDAR data. Some cyclist detection algorithms exist that can possibly fulfill this requirement [ 21 , 32 ]. In this research, the data of the cyclist’s area is manually cropped from the data collected by LiDAR sensor and used for orientation estimation.…”

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

“…This research also assumes that cyclists have been detected by other methods. For example, it is possible to convert a 3D point cloud into RGB-map in Bird’s Eye View (BEV) and implement image object detection algorithm, YOLO, on the BEV image [ 21 ]. Following this idea, the point cloud data of the cyclist is segmented from BEV in the software CloudCompare.…”

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

“…Abadi et al proposed to estimate the cyclist head and body orientation using joined head map information generated from Openpose [ 18 ], and then used the joined head and body orientation to predict the crossing intention of cyclist [ 19 , 20 ]. In order to identify the cyclist heading and predict their intentions, Garcia et al proposed a multi-class detection with eight classes according to orientations and presented a performance comparison for cyclist detection and orientation classification between the main deep-learning-based algorithms reported in the literature, such as SSD, Faster R-CNN and R-FCN [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cyclist Orientation Estimation Using LiDAR Data

Chang

Goncharenko

et al. 2023

Sensors

View full text Add to dashboard Cite

It is crucial for an autonomous vehicle to predict cyclist behavior before decision-making. When a cyclist is on real traffic roads, his or her body orientation indicates the current moving directions, and his or her head orientation indicates his or her intention for checking the road situation before making next movement. Therefore, estimating the orientation of cyclist’s body and head is an important factor of cyclist behavior prediction for autonomous driving. This research proposes to estimate cyclist orientation including both body and head orientation using deep neural network with the data from Light Detection and Ranging (LiDAR) sensor. In this research, two different methods are proposed for cyclist orientation estimation. The first method uses 2D images to represent the reflectivity, ambient and range information collected by LiDAR sensor. At the same time, the second method uses 3D point cloud data to represent the information collected from LiDAR sensor. The two proposed methods adopt a model ResNet50, which is a 50-layer convolutional neural network, for orientation classification. Hence, the performances of two methods are compared to achieve the most effective usage of LiDAR sensor data in cyclist orientation estimation. This research developed a cyclist dataset, which includes multiple cyclists with different body and head orientations. The experimental results showed that a model that uses 3D point cloud data has better performance for cyclist orientation estimation compared to the model that uses 2D images. Moreover, in the 3D point cloud data-based method, using reflectivity information has a more accurate estimation result than using ambient information.

show abstract

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cyclist Orientation Estimation Using LiDAR Data

Chang

Goncharenko

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…In [196], thermal imaging, enabling day/night time and illumination-independent data collection, was used to implement a robust pedestrian and cyclist detection using the Faster R-CNN [122]. Details of other related approaches are available in [197], [198], and [199].…”

Section: ) Detection Typesmentioning

confidence: 99%

Toward Performing Image Classification and Object Detection With Convolutional Neural Networks in Autonomous Driving Systems: A Survey

Turay

Vladimirova

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Williams [25] innovated a mobile app that exploits the internal sensors of an iOS smartphone to detect a bicycle crash using a threshold-based detection algorithm with the assistance of Apple's CMMo-tionManager sensor fusion algorithms, and when a crash is recognized, automatically invoke the virtual voice assistant Amazon Alexa to guide the cyclist through the crash reporting procedure. Aside from the tech-aided facilities applied to the cyclist's side, there are also solutions designed for other vehicles on the road or the ITS, such as a vehicle open door safety system proposed by Zhu et al [26], an automatic traffic monitoring and management system for pedestrians and cyclists developed by Pourhomayoun [27], and a cyclist orientation detection algorithm presented by Garcia-Venegas et al [28], most of which exploit the technologies of computer vision and machine learning to identify cyclists or their moving statuses. In [29], Alvi et al provided us with a thorough survey on varied IoT-based traffic accident detection systems -though not specifically focusing on cycling -using different kinds of sensors along with various sorts of detection algorithms, among which the machine learning technology is seemingly the most dominant tool used by researchers over the past decade.…”

Section: Introductionmentioning

confidence: 99%

Development of a cycling safety services system and its deep learning bicycle crash model

et al. 2022

View full text Add to dashboard Cite

This research developed an Internet of things (IoT) services system for cyclists to keep them safe when cyclingfrom terminal to back end -called the cycling safety services system. The proposed system consists of wearable devices, a service mobile app, and back-end services, primarily providing three categories of services: (1) The cycling team services, (2) the physiological status services, and (3) the environmental information service, incorporating the technologies of deep learning, IoT end device development, mobile app programming, RESTful API implementation, open data exploitation, etc. The proposed system aims at protecting the cyclists from being left out when cycling as a team, warning them when their physiological status is going to be abnormal or when there is a possibility of a bicycle crash, as well as proactively sending out urgent messages with location information when a crash occurs. Moreover, to enable the mobile app to recognize crash events, this research trained a deep learning bicycle crash model of 87.8049% accuracy and implemented a procedure in the mobile app based on this model to detect bicycle crashes automatically. The proposed system also provides a way for the cyclists to report any false crash alarm so that the crash model can be re-trained to reduce its false alarm ratio after the system is distributed to the consumers in the future. This research confirmed that the proposed system works well, and suggests that the proposed system, the crash model, the data collection method with its associated mobile apps, and the anonymous crash locations collected in the future can be valuable and contribute to the cycling society and relevant researchers in a positive way.

show abstract

On the safety of vulnerable road users by cyclist detection and tracking

Cited by 10 publications

References 47 publications

Cyclist Orientation Estimation Using LiDAR Data

Cyclist Orientation Estimation Using LiDAR Data

Toward Performing Image Classification and Object Detection With Convolutional Neural Networks in Autonomous Driving Systems: A Survey

Development of a cycling safety services system and its deep learning bicycle crash model

Contact Info

Product

Resources

About