GreenMap: Urban Greenery Mapping

EcoVision harnesses cutting-edge AI to create detailed maps of urban greenery from aerial imagery, promoting sustainable urban planning and enhancing community well-being.

Comment

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for GreenMap: Urban Greenery Mapping

About the Project: EcoVision Inspiration The inspiration for EcoVision came from the increasing need for sustainable urban development. As cities grow, green spaces are often overlooked, leading to negative impacts on public health and the environment. I wanted to leverage the power of AI to map and analyze urban greenery, helping city planners and communities make informed decisions about green space management.

Learning Experience Throughout this project, I deepened my understanding of:

Foundation Models: Gained hands-on experience with state-of-the-art models like Vision Transformers for image segmentation tasks. Data Handling: Improved skills in preprocessing aerial imagery and managing large datasets for training and testing. Real-World Impact: Learned about the importance of data visualization and its role in communicating complex information to stakeholders. Project Development Model Selection: I chose the Vision Transformer due to its effectiveness in handling high-resolution images and capturing intricate details. Data Preparation: Utilized provided datasets, ensuring to preprocess images for optimal model training. I created labeled datasets to train the model on different types of greenery. Fine-Tuning: The model was fine-tuned using the curated training data to enhance its accuracy in identifying various greenery types in urban settings. Evaluation: After training, I tested the model on a separate dataset, analyzing its performance using metrics like accuracy and F1 score. Visualization: Developed a user-friendly interface to visualize the output maps, allowing users to easily interpret the data. Challenges Faced Data Quality: Some aerial images had varying resolutions and lighting conditions, which affected model performance. I had to implement techniques for data augmentation to improve robustness. Model Complexity: Fine-tuning the model required careful tuning of hyperparameters, which was time-consuming but crucial for achieving desired accuracy. Visualization: Creating clear, informative maps from model outputs posed challenges in ensuring that the information was easily understandable to non-technical users. Conclusion EcoVision not only showcases the potential of foundation models in mapping urban greenery but also emphasizes the importance of sustainable urban planning. This project has equipped me with valuable skills and insights that I look forward to applying in future endeavors.

Share this project:

Updates