Finding Brick Kilns from Space: A Geovisual Search Pipeline
Abstract/Contents
- Abstract
- Brick production is central to economic development in Bangladesh. The industry is dominated by traditional brick kilns that generate substantial environmental and human health harms. Despite numerous regulations governing where brick kilns can be established and the type of manufacturing processes used, brick kilns are poorly regulated and tracked by the government. A lack of objective data on how many kilns exist, what type of technology is used, and where they are located inhibits ef- fective governance. Yet obtaining this type of data is difficult given the low-capacity of the government. We address these limitations by taking advantage of advances in machine learning and satellite imagery to develop a scalable and replicable geovisual search pipeline to visually identify all of the brick kilns in Bangladesh. By taking a weakly supervised approach to localization, we create a model that detects kilns using only sparse classification data. We also introduce an approach for scaling that can significantly improve the efficiency of models run on satellite imagery.
Description
| Type of resource | text |
|---|---|
| Date created | May 28, 2019 |
| Publication date | May 16, 2023 |
Creators/Contributors
| Author | Lee, Jihyeon Janel |
|---|---|
| Contributor | Brooks, Nina |
| Research team head | Ermon, Stefano |
| Thesis advisor | Lobell, David |
| Thesis advisor | Burke, Marshall |
| Thesis advisor | Luby, Steve |
Subjects
| Subject | computer science |
|---|---|
| Subject | machine learning |
| Subject | computer vision |
| Subject | sustainability |
| Subject | health |
| Genre | Text |
| Genre | Thesis |
Bibliographic information
| Related item |
|
|---|---|
| Location | https://purl.stanford.edu/xc555rn1814 |
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 4.0 International license (CC BY-NC).
Preferred citation
- Preferred citation
- Lee, Jihyeon Janel and Brooks, Nina and Ermon, Stefano and Lobell, David and Burke, Marshall. (2018). Finding Brick Kilns from Space: A Geovisual Search Pipeline. Stanford Digital Repository. Available at: https://purl.stanford.edu/xc555rn1814
Collection
Undergraduate Theses, School of Engineering
View other items in this collection in SearchWorksContact information
- Contact
- jihyeon@cs.stanford.edu
Also listed in
Loading usage metrics...