The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.
Dataset Card for Dataset Name
Images accessibility infrastructure Rotterdam.
This dataset card aims to be a base template for new datasets. It has been generated using this raw template.
Dataset Details
Dataset Description
This dataset consists of images pertaining the mobility obstacles in the city of Rotterdam. It consists of both positive and negative influences on the accessability of public space to those who suffer from either mobility, or sight issues.
- Curated by: Anja Ellwood, Benedict Heuff, Eilean Dewald, Marit de Hoogh, Vera Hoekstra
- Language(s) (NLP): English
- License: N/A
Dataset Sources [optional]
Uses
The dataset was compiled in order to train a convolutional neural network model to identify problem areas affecting mobility within the city of Rotterdam.
Direct Use
This dataset can be used to train models on mobility issues within Dutch cities, or to examine how mobility challenges are tackled within this context.
Out-of-Scope Use
Out-of-Scope uses for this dataset include any and all purposes not directly related to urban mobility infrastructure.
Dataset Structure
the data is split in two sections. Good, which consists of images containing accessible public spaces, and Not_good which consists of images containing inaccessible or difficult to access public spaces
Dataset Creation
Curation Rationale
This project aimed to contribute to the mapping of accessibility in Rotterdam by collecting images of various accessibility features and barriers in public spaces.
Source Data
The data was manually collected by the curators, through in person photography, and the datascraping of Google maps.
Data Collection and Processing
The data collected had to showcase city infrastructure or features that impacted at least one form of mobility. Furthermore, images collected had to be of public spaces.
Who are the source data producers?
Anja Ellwood, Benedict Heuff, Eilean Dewald, Marit de Hoogh, Vera Hoekstra. All of these producers are able-bodied Europeans
Personal and Sensitive Information
All personal, sensitive, or private data has been scrubbed from the dataset.
Bias, Risks, and Limitations
There is bias present in this dataset. It has been collected through two methods, both in person photography, and datascraping, both of these methods have been coloured by the perception of able-bodied individuals.
Recommendations
Due to time and technological constraints this dataset remains rather limited, which should be taken into account when using it to train any model.
- Downloads last month
- 5