Our inspiration was that we wanted to target something that affected the majority of the world's population. That was it would be really impacting.
What it does
Our app, after being completely developed, will allow the user to take a picture of any item they want to dispose of, and then our computer will determine how to dispose of it correctly. As of right now, our photo recognition is nearly developed, but as a backup, we have a survey that allows the user to get the same result, but unfortunately with a little more work.
How we built it
We used tensor flow and implemented machine learning to teach the computer about different waste items. As of now, we have 73 percent accuracy. We used Html and CSS to create the survey. We originally used java to make this, but then remade it using HTML and CSS to make it more user-friendly.
Challenges we ran into
Some challenges we ran into were that we were unable to resize images to use to make our machine learning accuracy better. We also faced challenges in how to implement HTML and CSS to recreate the form as none of us have much experience in those coding languages.
Accomplishments that we're proud of
We got the tensor flow photo identification to work. Well, the machine learning part. It was very complex as we had absolutely no idea about the topic. We also didn't know anything about HTML or CSS but still managed to make a functional program using them.
What we learned
We learned about machine learning using tensor flow, we learned about how to make graphical games using java, and we learned a lot about HTML and CSS
What's next for Upcycle
We want to make our photo recognition software more advanced and accurate and implement this into our app. We have an app, but unfortunately, with the time being given, we weren't able to develop our app as much as we'd like too. We would also like to add an advanced feature that allows people all around the world to communicate about beach clean-ups, protests and more.