We wanted to use Telus IoT network to support farmers who work in rural and often isolated areas. In addition there are many towns where there is only one grocery store making fresh produce inaccessible to many people.
What it does
AuTomato is an IoT machine that enables consumers to buy fresh produce that has been stocked by farmers at accessible locations.
How we built it
We used the Telus IoT to supply Cellular data for the AuTomato machines that will be placed in remote locations. We trained a machine learning model using computer vision to detect the freshness of food.
Challenges we ran into
We found it difficult to integrate all the different components together. We were fairly new to computer visiion and the RaspberryPis. We also found it hard to convert the data sent in bytes from the camera into ASCII.
Accomplishments that we're proud of
We are able to build and train a deep neural network model that can detect and distinguish between a fresh apple and a rotten one with high accuracy. We implemented various sensors to collect data for needs analysis for local produce and allow access to fresh produce in remote communities and for farmers to be able to sell their produce in a wider range of areas aside from just farmers markets.
What we learned
We learn a lot about how to integrate multiple aspects of a system utilizing databases and cloud storage. We also learned about the architecture of a convolution neural network and about the various communication protocols such as I2C and Serial, which we integrated into this project.
What's next for AuTomato
We hope to incorporate more functionality into our project that will allow farmers to monitor the stocks of each AuTomato machine in real time.