Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2407.01117 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909235614318592 |
|---|---|
| author | S, Eliza Femi Sherley T, Sanjay P, Shri Kaanth S, Jeffrey Samuel |
| author_facet | S, Eliza Femi Sherley T, Sanjay P, Shri Kaanth S, Jeffrey Samuel |
| contents | This article includes a comprehensive collection of over 800 high-resolution streetlight images taken systematically from India's major streets, primarily in the Chennai region. The images were methodically collected following standardized methods to assure uniformity and quality. Each image has been labelled and grouped into directories based on binary class labels, which indicate whether each streetlight is functional or not. This organized dataset is intended to make it easier to train and evaluate deep neural networks, allowing for the creation of pre-trained models that have robust feature representations. Such models have several potential uses, such as improving smart city surveillance systems, automating street infrastructure monitoring, and increasing urban management efficiency. The availability of this dataset is intended to inspire future research and development in computer vision and smart city technologies, supporting innovation and practical solutions to urban infrastructure concerns. The dataset can be accessed at https://github.com/Team16Project/Street-Light-Dataset/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_01117 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Comprehensive Dataset for Urban Streetlight Analysis S, Eliza Femi Sherley T, Sanjay P, Shri Kaanth S, Jeffrey Samuel Computer Vision and Pattern Recognition This article includes a comprehensive collection of over 800 high-resolution streetlight images taken systematically from India's major streets, primarily in the Chennai region. The images were methodically collected following standardized methods to assure uniformity and quality. Each image has been labelled and grouped into directories based on binary class labels, which indicate whether each streetlight is functional or not. This organized dataset is intended to make it easier to train and evaluate deep neural networks, allowing for the creation of pre-trained models that have robust feature representations. Such models have several potential uses, such as improving smart city surveillance systems, automating street infrastructure monitoring, and increasing urban management efficiency. The availability of this dataset is intended to inspire future research and development in computer vision and smart city technologies, supporting innovation and practical solutions to urban infrastructure concerns. The dataset can be accessed at https://github.com/Team16Project/Street-Light-Dataset/. |
| title | Comprehensive Dataset for Urban Streetlight Analysis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.01117 |