BirderBase

Inspiration

Birderbase makes learning about birds easy and fun, by sharing information on the local birdwatching spots. It makes bird watching more accessible to all the avians of Philadelphia. Our team, united by a shared enthusiasm for ecology and technology, was inspired to bridge the gap between nature enthusiasts and technological advancements. Birderbase was conceived as a solution to enrich the birdwatching experience, allowing users not just to observe but to engage deeply with their environment through data-driven insights. We envisioned a platform where technology enhances nature exploration, making it more accessible, informative, and interactive for everyone from amateur birdwatchers to seasoned ornithologists. This project reflects our commitment to fostering a deeper connection between people and the wildlife around them, empowering them with knowledge and tools to appreciate and protect our feathered friends.

What it does

BirderBase is an interactive web application that uses advanced mapping and computer vision algorithms to enhance the birdwatching experience. This single-page application enables users to explore clusters of bird sightings around Philadelphia, offering an intuitive map-based interface for scouting potential birdwatching locations.

Upon visiting BirderBase, enthusiasts can immediately begin their exploration of nearby bird habitats. For a more interactive experience, users can upload images of birds they've encountered during their birdwatching expedition. Leveraging the power of Microsoft Azure, our advanced Computer Vision Analytics processes these images to generate relevant tags, helping to pinpoint the species observed.

Further enhancing user engagement, our API—powered by OpenAI—provides comprehensive descriptions of each bird, striving for precise identification. Our robust computer vision model is specifically designed to recognize birds from considerable distances, making it an invaluable tool for birdwatchers seeking to deepen their knowledge and enhance their field experiences.

How we built it

BirdSpot was developed during a 24-hour hackathon by a team of three. We utilized the following technologies: NodeJS ExpressJS Azure Computer Vision Firebase Leaflet.js LLM

Challenges we ran into

Throughout the development of BirderBase, our team faced several significant challenges. The primary hurdle was seamlessly integrating diverse functionalities into the website. Specifically, we struggled to effectively display the results from our computer vision predictions alongside the AI-generated textual descriptions directly on the main website interface. This integration issue was critical as it affected the core user experience, hindering our ability to provide a smooth and cohesive interaction for birdwatchers utilizing our platform.

Accomplishments that we're proud of

In the intense, round-the-clock environment of our development cycle, we achieved several significant milestones that we're extremely proud of. Notably, we successfully implemented and operationalized three distinct AI-driven features on our website within just 24 hours. Our mapping integration, a critical component of the user experience, functions flawlessly, providing users with a seamless and intuitive way to locate bird sightings. Furthermore, we excelled in the realm of image recognition—our system can now accurately identify various bird species from user-uploaded images. These accomplishments reflect our team's dedication and technical prowess in creating a tool that birdwatchers will find invaluable.

What we learned

During the development of BirderBase, our team deepened our understanding of integrating Azure-powered machine learning tools and managing data with MongoDB. This experience enhanced our skills in utilizing cloud ML technologies and NoSQL databases, equipping us to handle complex data scenarios and improve application performance efficiently.

What's next for BirderBase

In the future, we aim to enhance the image recognition system to be more accurate by making the computer vision model specify the exact species of the bird.