WaterCV

Our computer vision program helps wastewater treatment plant operators make data driven decisions by converting visual information into objective and quantitative data.

Comment

Inspiration

Wastewater treatment is a mechanically and biologically intensive process and wastewater treatment operators are responsible for managing these systems. There are times that an operator has to rely on their own eyes to tell them how a process is performing that is subjective and the interpretation can vary from operator operator. One example is determining if the biology of a tank is healthy based on the color of the water and the amount of foam accumulated at the top. The amount of foam, color of foam, and color of the water are all parameters that are used to make changes to the most energy intensive part of a wastewater treatment process. Money is wasted every year based on how one operator interprets a color they see in this process. We wanted to see if we could create a tool that would remove this subjectivity from the process and allow operators to make decisions based on objective data.

What it does

We use computer vision to analyze a camera video feed of the surface of a biological treatment process and quantify the amount of foam, the color of the foam, and the color of the water under the foam. This data is then displayed in a convenient web page with tools for analysis.

How I built it

Our computer vision developer used C++ and OpenCV to analyze video footage of the water and output numerical data. Our web developer created a web page with HTML, CSS, and Javascript to display and analyze the data.

Challenges I ran into

We had to combine the two sides of our project in a few hours, with one computer running Linux and the other running Windows.

Share this project:

Updates