Techolocation

Inspiration

Last year, one of our team members was unknowingly hiking on a path with an active cougar. If she hadn't been lucky, she could have gotten severely injured or died because she was not alerted of the dangers on the trail. This made her and all of us acutely aware of the imbalance that climate change and urbanization causes on structural diversity in our fast-paced world, and how if we do not curb this imbalance now, future generations may never get to experience the beauty of nature.

What it does

TechoLocation is a mobile app that uses image recognition technology and crowdsourced data to detect and map invasive and dangerous species in real-time. Users can snap photos of plants or animals, and the app instantly identifies whether they pose a threat, alerting users with safety guidelines. The app also tracks sightings on a map, helping communities stay informed about local wildlife hazards.

How we built it

We built Techolocation using React Native for cross-platform compatibility, allowing both iOS and Android users to access it. We integrated Teachable Machine AI for image recognition to identify species, and Leaflet API to plot sightings in real-time. The back end is powered by cloud-based storage, where user-submitted data is processed and stored for ongoing monitoring and analysis.

Challenges we ran into

One of our biggest challenges was the time zone difference, which meant we had to start working at 4 AM on both days. Nevertheless, we overcame this roadblock by sleeping early, (some coffee and lots of naps), and staying motivated by choosing a topic that we are all passionate about. Furthermore, we had a lot of difficulties learning unfamiliar concepts quickly, such as using React to develop our app and incorporating different APIs, by doing thorough research and asking for help when we needed it.

Accomplishments that we're proud of

We’re proud of successfully building a functional app that can not only identify species but also alert users in real-time. Our ability to integrate image recognition with mapping features was a huge milestone, since none of us had focused on this aspect of programming before. We are also very proud of not giving up even when every aspect of our project seemed to fail, and in the end, we are very grateful that we were able to incorporate everything seamlessly.

What we learned

We learned a lot about the importance of data accuracy and how essential it is for training machine learning models. We also gained valuable insight into UI design with javascript, realizing how critical it is to create an app that’s both powerful and simple to use. Furthermore, we learned a lot about different databases and how to do effective research to find the right ones. Beyond technical knowledge, we learned about the real-world impact of invasive species and the importance of engaging communities in conservation efforts.

What's next for Techolocation

Our next steps for Techolocation include fully integrating the machine learning feature directly into the app to streamline the species recognition process. We also plan to expand our database by adding more species, both invasive and dangerous, to ensure comprehensive coverage. Updating the recognition system to be more accurate and effective is a priority, as is opening up the app to the public to crowdsource more data and maximize our impact. Additionally, we want to introduce educational content within the app, helping users learn about the environmental issues we’re tackling and empowering them to take meaningful action.

Built With

• apis
• css
• databases
• html
• image-recognition
• javascript
• leaflet.js
• machine-learning
• react
• react-native
• teachable-machine
• typescript