trapedesveh-a-mini-dataset-for-intelligent-transportation-systems
This repository contains a collection of labelled images of vehicles, pedestrians and street signs.
https://github.com/judith989/trapedesveh-a-mini-dataset-for-intelligent-transportation-systems
Science Score: 57.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 1 DOI reference(s) in README -
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (9.6%) to scientific vocabulary
Repository
This repository contains a collection of labelled images of vehicles, pedestrians and street signs.
Basic Info
- Host: GitHub
- Owner: Judith989
- License: cc0-1.0
- Default Branch: main
- Size: 44.2 MB
Statistics
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
TraPedesVeh
This repository contains the TraPedesVeh mini dataset which is a collection of annotated images containing vehicles, traffic signs and pedestrians. It was first introduced in the paper titled "State-of-the-Art Object Detectors for Vehicle, Pedestrian, and Traffic Sign Detection for Smart Parking Systems", which was presented at the International Conference on Information and Communication Technology Convergence (ICTC2022), Jeju, South Korea.
Dataset Description
The dataset was built by collecting random images from the internet, using the keywords: traffic signs, vehicles and pedestrians. A total of 230 different images were collected. The annotation of images was done manually using a graphical anotation tool named Labelimg. As illustrated in the table below, the annotation process produced a total of 938 annotated objects, for 7 different classes of objects. The Labelimg program generates an XML file for each of the images annotated.
The train folder contains 146 annotated images and XML files, and the test folder contains 46 annotated images and XML files.
Three other files are included in this repository: test.tfrecord contains the test images, train.tfrecord contains the training images, and label_map.pbtxt contains the labels. These will be very useful when training on object detection models that support tfrecord formats.
The Tfrecord files were results from an augmented version of the original dataset. As a result, the training tfrecord file contains 501 images, while the test file contains 61 images.
Abstract
Paper
Authors
Judith Nkechinyere Njoku, Goodness Oluchi Anyanwu, Ikechi Saviour Igboanusi, Cosmas Ifeanyi Nwakanma, and Dong-Seong Kim
Citation
If you use this dataset in your research, please cite this repository. or cite this paper:
"J. N. Njoku, G. O. Anyanwu, I. S. Igboanusi, C. I. Nwakanma, J. -M. Lee and D. -S. Kim, "State-of-the-Art Object Detectors for Vehicle, Pedestrian, and Traffic Sign Detection for Smart Parking Systems," 2022 13th International Conference on Information and Communication Technology Convergence (ICTC), 2022, pp. 1585-1590, doi: 10.1109/ICTC55196.2022.9952856."
License
Creative Commons Zero v1.0 Universal Link
Owner
- Name: Judith Nkechinyere Njoku
- Login: Judith989
- Kind: user
- Website: judithnjoku.com
- Repositories: 3
- Profile: https://github.com/Judith989
Machine Learning Engineer | Writer | Researcher | Metaverse Enthusiast
Citation (CITATION.cff)
# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!
cff-version: 1.2.0
title: >-
TraPedesVeh-A-mini-Dataset-for-Intelligent-Transportation-Systems
message: >-
If you use this dataset, please cite it using the
metadata from this file.
type: dataset
authors:
- given-names: 'Judith Nkechinyere'
family-names: Njoku
email: judithnjoku24@gmail.com
affiliation: Kumoh National Institute of Technology
orcid: 'https://orcid.org/0000-0002-2294-9204'
- given-names: 'Goodness Oluchi '
family-names: Anyanwu
email: anyanwu.goodnes@kumoh.ac.kr
orcid: 'https://orcid.org/0000-0002-0304-4143'
- given-names: 'Ikechi Saviour '
family-names: Igboanusi
email: ikechisaviour@kumoh.ac.kr
affiliation: Kumoh National Institute of technology
orcid: 'https://orcid.org/0000-0003-2325-5105'
- given-names: 'Cosmas Ifeanyi'
family-names: Nwakanma
email: cosmas.ifeanyi@kumoh.ac.kr
affiliation: Kumoh National Institute of technology
orcid: 'https://orcid.org/0000-0003-3614-2687'
- given-names: 'Dong-Seong'
family-names: Kim
email: dskim@kumoh.ac.kr
affiliation: Kumoh National Institute of technology
orcid: 'https://orcid.org/0000-0002-2977-5964'