It is crucial to understand the quality of a street view imagery (SVI) dataset to assess its ’fitness for purpose’. In this repository, we present: (1) a Jupyter notebook to quickly assess the quality of a SVI dataset by 9 quality elements, each at its relevant hierarchical levels - image, street, or grid; (2) a sample dataset for running the notebook.
A total of 9 quality elements are evaluated in the notebook with their spatial variations visualised:
- Spatial coverage
- Spatial continuity
- Count
- Age of the most recent coverage
- Age of the first available coverage
- Number of years covered
- Number of months covered
- Time elapsed between coverage
- Image blurriness
For demonstration, we provide a set of Mapillary SVI data from a 2 km x 2 km area in Kowloon, Hong Kong for running the notebook. This includes:
- A metadata file, obtained from the Mapillary API
- A zip file of sample images, obtained from Mapillary
Apart from the SVI data, we also provide a raster file that enables the grid-based analysis, obtained from WorldPop.
All data in the sample dataset was obtained under the CC BY-NC-SA 4.0 license.
The expected output for each quality element, if the sample dataset is used:
- Spatial coverage
- Spatial continuity
- Count
- Age of the most recent coverage
- Age of the first available coverage
- Number of years covered
- Number of months covered
- Time elapsed between coverage
- Image blurriness
The sample dataset can be downloaded from Google drive.
A paper about the work was published in the International Journal of Applied Earth Observation and Geoinformation and it is available open access.
If you use this work in a scientific context, please cite this article.
Hou Y, Biljecki F (2022): A comprehensive framework for evaluating the quality of street view imagery. International Journal of Applied Earth Observation and Geoinformation, 115: 103094. doi:10.1016/j.jag.2022.103094
@article{2022_jag_svi_quality,
year = {2022},
title = {{A comprehensive framework for evaluating the quality of street view imagery}},
author = {Hou, Yujun and Biljecki, Filip},
journal = {International Journal of Applied Earth Observation and Geoinformation},
doi = {10.1016/j.jag.2022.103094},
pages = {103094},
volume = {115}
}
This dataset is released under the CC BY-NC-SA 4.0 license.
Feel free to contact Yujun Hou or Filip Biljecki should you have any questions. For more information, please visit Urban Analytics Lab, National University of Singapore.
We appreciate the valuable contributions of the VGI community.
We thank our colleagues at the NUS Urban Analytics Lab for the discussions.
This research is part of the project Large-scale 3D Geospatial Data for Urban Analytics, which is supported by the National University of Singapore under the Start Up Grant R-295-000-171-133.