We love old people

What it does

Our pill identifier allows a user to take a picture or ipload a file of a pill and in return get which pill is most likely classifies as from our data set.

How we built it

We used a machine learning algorithm as well as a neural network model that would allow for images to be compared against our dataset. The model would look for similarities in shape, size, and color, and return the name of the pill that most accurately represents input as well as the prediction percentage that tells the user how accurate the model is.

Challenges we ran into

  • Fine tuning the neural network
  • Implementing camera functionalities with the javascript
  • Linking the front-end to the back-end ## Accomplishments that we're proud of Successfully implementing a server that runs a decent neural network Successfully making a user friendly interface that can take video input. ## What we learned Making a neural network is extremely complex ## What's next for Pill Identifier Linking the front end and the back end as well as fine tuning the nural network.
Share this project: