Finding Nemo

Home Page
App Manual
App Manual
App Manual
Police Interfact
User Interface
Location access- User Interface
Results of Match found- User Interface
Email notification

Inspiration

Every year, over a million children are reported missing around the world. According to the Federal Bureau of Investigation, NCIC [1], every year, 460,000 children are reported missing in the United States alone. This, however, is only a fraction of the number of children who are actually missing; many missing cases aren’t even reported, and the database is also not very accurate for a few countries. Especially in countries like India, amber alerts aren’t very popular, and with its large population, it becomes very hard to trace abduction cases. In this project, we try to work on resolving the real-life problems people and police face while searching for a missing child.

What it does

With the help of this app, a person can scan the face of a child whom that person believes to be a potentially missing child. Our app will compare the scanned face to the database of lost children and send an alert email to the police department and the family of the child regarding the location of the child if a match is found. This app will not only help in locating lost children more efficiently, but it will also help the police to locate sources of child abductors, thereby reducing the cases of child abduction.

How we built it

The app will use Flutter [2] for the frontend, and the backend and data will be handled by Firebase [3]. We intend to use the Google Facenet API, which offers us pre-trained machine learning models for our face recognition feature. The email service used by this app is intended to be EmailJS [4]. The following is a brief summary by means of the following architecture diagram of how the app will function: https://drive.google.com/file/d/1nWRwLdZlLFl80_af0Y7xBPYXpAvWJTuh/view?usp=sharing

Challenges we ran into

The biggest challenge we faced was setting up the Regula Face API instead of having a TensorFlow Lite model integrated in the app itself.

Accomplishments that we're proud of

We take pride in creating an app that, in some lmanner, contributes to society and helps parents in finding their lost children.

What's next for Finding Nemo

A potential upgrade we foresee for this project is replacing Regula API with something like TensorFlow Lite model and increase some security features.