Images

6330 images

Explore Dataset

Overview

This Roboflow dataset composes of many synthetic and edge-case angles for cyclist detection. This dataset was created by Charles Tang, a high school student at the Massachusetts Academy of Math and Science located at WPI. This dataset will be used in the project A Blind Spot Alert Apparatus for Cyclists in Right-Turning Semi-trailer Trucks, which will be used to provide further object detection training in addition to the CIMAT cyclist dataset for the blind spot cyclist detection apparatus.

Abstract In order to reduce the number of truck-cyclist accidents, this study designs a visual-based blind spot warning system for a semi-trailer truck. A large number of cyclist collisions are caused by semi-trailer trucks, more specifically, in right-hook turns. First, the object detection model was created using state-of-the-art lightweight deep learning architectures trained on a cyclist image dataset, which is used to locate and detect cyclists actively. Next, the object detection model was deployed onto a Google Coral Dev Board mini-computer with a camera module and analyzed for accuracy and speed. Lastly, the combined blind spot detection device will be tested in real-time to model traffic scenarios and analyzed further for performance, feasibility, and further work of the apparatus.

Keywords: Bicyclist safety, blind spots, vulnerable road users, object detection, semi-trailer trucks, right-hook turns, cyclist collisions, warning system

Sources of Images

Parallel Domain synthetic bicycle detection dataset
Other Roboflow cyclist sources
Online digital sources

Please contact me at ctang5@wpi.edu for any questions or concerns.

Trained Model API

This project has a trained model available that you can try in your browser and use to get predictions via our Hosted Inference API and other deployment methods.

Cite This Project

If you use this dataset in a research paper, please cite it using the following BibTeX:

@misc{
                            bike-detect-ct_dataset,
                            title = { Bike Detect - CT Dataset },
                            type = { Open Source Dataset },
                            author = { Bicycle Detection },
                            howpublished = { \url{ https://universe.roboflow.com/bicycle-detection/bike-detect-ct } },
                            url = { https://universe.roboflow.com/bicycle-detection/bike-detect-ct },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2023 },
                            month = { feb },
                            note = { visited on 2024-04-24 },
                            }