PassingBy

Our solution helps industries that in need of special purification and equipment requirements to verify that every employee has fully worn their equipment which ensures our safety and product quality.

Comment

Inspiration

So our initial inspiration was our colleagues at INSAT where they get to study chemistry and expose themselves to certain dangerous chemicals, where sometimes they are left in danger and with no motivation to continue in the field, in a field where innovation takes place and humanity's history gets to be written. In addition to that, we were personally touched by one of our ex-Insatien's story, where she was left to die in a laboratory after being exposed to intoxicating chemicals. After we heard this story we decided to take action and make sure that we benefit both the industry investors and workers.

What it does

Our solution consists of providing a solution for industries that are requiring a special uniform/ensemble to be worn before entering a purified chamber. It is a camera that detects if the person entering is wearing his uniform/ensemble properly and by that, provides access to the mentioned.

How we built it

For our solution, the simplicity of the idea didn't require much hardware to implement it. All we needed was: RFID sensor, a camera, and for the micro-controller, a raspberry pi 4 v2. But also we needed an AI model that works on computer vision and identifies people with uniforms and whether to give them access or not. Also, a web platform to visualize our data that's composed of a front-end, back-end and a database.

Challenges we ran into

The first challenge we ran into was the configuration of the Rasberry pi. Also, implementing the messaging between our AI model and our raspberry pi card, in other words, the communication between TensorFlow and TensorFlow lite because we couldn't use TensorFlow library in the raspberry card. In addition to that, the data communication between the card and the database, it wasn't feasible at first but then with some research, it was achieved. Concerning the software part, ReactJs wasn't easy to implement with charts as it was getting into infinite for loops that were preventing it from transmitting data or showing graphics. After some time coding and researching it was achieved. Finally, the AI model that was changed multiple times, because at first we faced so many challenges concerning the database and data which wasn't found online. Also, applying transfer learning was achievable in a short amount of time so we had to switch to a different model locally on our computers which were hard to configure libraries on, especially Tenserflow.

Accomplishments that we're proud of

We are proud first of our resilience to change, since so many obstacles presented themselves at the very first moments of the challenge, we discussed our ideas openly and we were welcoming to any idea that each one us expressed. The ability to work as a solid team and support each other when some of us got down or were tired was certainly a win. In addition to that, what we learned because some of us had no idea at first about the industry or even industry 4.0 but thanks to this 48h challenge, we enriched our knowledge concerning both topics. Being in this ambiance and meeting new people, living this new experience. We are most honored and proud to have made it to this hackathon.

What we learned

We learned to strengthen our bonds and how to connect with people, and we learned how to deal with stressful accidents and last-minute bugs with less stress as possible. We learned how to deal with stress and how to manage our time, how to find innovative ideas, and keep an open mind for any passing thought. We learned to trust each other and listen to each other, and especially understand each other, for us being part of this team sent us to be friends and teammates for future hackathons. We learnt how to understand other cultures and make entertaining conversations with other team members and of-course we learned to have fun and enjoy ourselves.

What's next for PassingBy

Thank you for asking this question, there's so much in the future for PassingBy, the first being hopeful in winning this makeathon and raise a sum of money to start funding our idea, talking with possible investors and if there is a possibility, see if one of the sponsors is willing to act as an incubator to our project so we can start producing and selling it, and looking for markets for it, first in Tunisia and then internationally. But that's not everything. Since 48h are only enough to build a prototype, there are still so many ideas and future features to implement, from using multiple cameras to detecting different ensembles and forecasting for costs that we are preventing when using our solution.

Built With

Share this project:

Updates