Inspiration

One of our groups member's Aba walked into this challenge with a conspicuous childhood memory: the day her family's small store was obliterated by hurricane Harvey. Facing the long journey ahead, her family struggled as they had no insurance. For their small business they lost $120,000 in capital and goods. They were unable to find any reimbursements and fought a long battle to recover what could have been restored by proper access to insurance information.

What It Does

Our concept, H2OOPS: Splashing into Flood Coverage, focuses on free insurance counseling for small businesses in regards to business proneness to natural water disasters. User's input their business' location, and in return receive a calculated likelihood of being flooded by nearby bodies of water based on multitude and proximity. Then, user's input a photo of their business and their room is scanned for objects in danger to water damage, and gives a suggested cost incurred for those products. Finally, users are recommended an insurance plan based upon their percentage threat toward flood disasters derived as a result of personalized flood risk and potential harm to their products.

How We Build It

We built this using Tkinter as a framework for a GUI because the majority of our code was in Python 3. A Google Maps API was used for finding nearby rivers by finding the longitudes and latitudes of nearby natural water landmarks and using a distance conversion to find all rivers at most 10 miles away. The object recognition uses ImageAI in Yolo in order to scan for valuables, deem what items are worth the most, and calculate an estimated incurred cost.

Challenges We Ran In To

Some challenges we ran into was working in our UI with such a short time frame. One of our teammates made a beautiful figma mockup for UI that we could convert to HTML, but we ran out of time trying to use it through flask to orchestrate the use of Python and HTML. Another issue we had was installing the packages on one of our macs. We had to quickly pivot to host our demo on one of the windows computers and work through transferring files collaboratively through git.

Our Accomplishments

We're proud of how well we worked as a team, conquering the distributed development part of this hackathon and completing so much with only three teammates. We also stood strong and persevered this weekend, not stopped by massive challenges and encouraged to work persistently to come up with a solution.

What We Learned

We all learned something new this weekend: Elijah learned how to use an API, Aba discovered how to use ImageAI and an object recognition model, and Nykaela took her figma skills to a whole another level.