PlantLife | Devpost

Inspiration

We conducted different lab-based experiment utilizing different plants to see how they respond in different changes in their environment. The results weren't supposed to vary that much but we were astounded at the results. We noticed most of the plants were dying and the rest of the plants were perfectly healthy. Curious about this, we examined the plants which had perished. We noticed fungi that had formed around the roots of one of the plants. As we examined the other plants, we noticed a general trend with more and more infectious bacteria that had formed around these plants. We came to the conclusion that the main leading factor to the death of these plants were due to the spread of diseases. Other factors we found were little water and less exposure to sunlight. From the perfectly healthy plants, this told us that the regular cycle we provided with these plants was working and was correct way in dealing with how to treat these of plants. However, for the perished plants if they never had been infected with diseases we concluded that they would also be healthy as well. Overall, this inspired us to create this app PlantLife to detect whether or not the plant is healthy or has a disease to offer treatment offers to help better protect the environment in many area, mainly in India.

What it does

PlantLife is essentially an AI app which takes a sample of any plant and produces a result, notifying the user if it's healthy or if it's infected with a disease. The fascinating thing about this app is if it's labeled as infected, the app will provide the user with what pest or disease the plant is infected with. Following with what disease the plant is infected with, this app gives the user with various treatment options, linking you to a website based on the type of disease identified.

How we built it

We built this app using X code. For the ML model, we used Jupiter and Google Collab.

Challenges we ran into

Some challenges we ran into for PlantLife were mainly directed to the ML model. We couldn't find a file or directory matching up with the image folder. Processing time was also really long and it was hard to import on Jupiter and Google Collab. The other issues were directed to the app where it was tedious to integrate machine learning on X code. It was also difficult process some softwares on this software.

Accomplishments that we're proud off

We're proud that PlantLife has been developed to being a well, adequate app which functions and works properly. This app can help save many plants by spreading more awareness with the treatment options, helping protect the environment in areas with poor vegetation in India. This can help make these places an eco friendly manner by creating a more greenhouse-gas area.

What we learned

We learned new exposure and gained more experience to different platforms and languages like swift, python, and GoogleCollab. This app has also helped us utilize different ideas using machine learning and artificial intelligence to help protect the environment. This can also help prepare us and make upcoming hackathons easier to approach as we are dealing with various topics and exposure which we can implement on to create new designs and advances in machine learning and artificial intelligence apps.

What's next for PlantLife

What we would want next for PlantLife is along with the treatment options, we would want them to be linked to a website where it can be accessible databased with machine learning. We would also want to add additional features to the app like with the app processing different treatment options for the plant, it would be really helpful if the app could find the nearest locations from the plant to a specific treatment center. This can be done using geospatial and navigational uses to develop a proximity radius and help spread awareness for the user to get the plant treated immediately.