Deep Green Diagnostics: Urban Green Space Analysis Using Deep Learning and Drone Images

Area of Science:

• Environmental Science
• Urban Planning
• Public Health

Background:

• Increasing global urban populations necessitate monitoring urban environmental health.
• Urban green space quality and quantity are linked to population health.
• Supervising large urban land areas for contamination is a significant challenge.

Purpose of the Study:

• To develop a deep learning solution for assessing urban land health and contamination.
• To create software for generating updated maps of land health and contamination in cities.
• To support public health institutions in urban environmental management.

Main Methods:

• Utilizing deep learning algorithms for land health assessment.
• Developing software for automated mapping of urban land conditions.
• Analyzing land data to identify potential contamination.

Main Results:

• A functional deep learning-based software for urban land health assessment.
• Experimental data demonstrating the software's capability in identifying land health issues.
• Freely available open-source code and experimental data.

Conclusions:

• Deep learning offers a viable solution for monitoring urban land health at scale.
• The developed software can significantly aid public health efforts in large cities.
• Open access to code and data promotes further research and application in urban environmental monitoring.