A

AgroMed

Field Level Identification of Pest and Diseases in Plants using Machine Learning Model. It predicts Pest or Disease infection in Plant Leaves and suggsets biological and chemical solutions for it.

A

AgroMed

Field Level Identification of Pest and Diseases in Plants using Machine Learning Model. It predicts Pest or Disease infection in Plant Leaves and suggsets biological and chemical solutions for it.

The problem AgroMed solves

Starting with the early crop stages, a farmer must closely monitor crops because of various crop insects, pests and diseases.
Depending on the crop type and growth stage, it's estimated that early pest detection can reduce yield loss by up to 20-40%.
Therefore, farmers need to put all of their effort into constant crop monitoring.
Through easy monitoring, a farmer can easily collect information about the presence of plant diseases through capturing images of the affected areas of the leaves.
Through immediate solutions they don't want to wait for any other professionals to detect the disease and to treat them.
The application suggests amount of Urea/MOP/SSP/DAP for the field relative to the total area of the field and the crop grown in it.
It also suggests chemical and biological solutions based on the plant disease identified.
Farmers can easily handle the problems with a smartphone and thereby increase the yield.

The agriculture department as a whole don't need to monitor every farms and fields physically.
By using the app every farmer is able to manage crops and its diseases on his own.
They even get a detailed report of the problems faced by each and every farmers in their locality,
like the kind of disease their plants are facing, the method they use to cure them. We give both biological and chemical pest control solutions for each problem detected.
The biological control minimizes the environmental , legal & public safety concerns.
This app can also provide agricultural departmental notifications directly to the farmers smartphone, details about collection of harvested crops by the government etc.

Challenges we ran into

There were a lot of bugfixes we did through the process of creating this project, at first the tfmodel we imported did not give us any response, but at later stages of testing it worked properly.
There was problems at the firebase end which we referred the internet for clarification.
Asset images were not showing up in the application, then we tried network images and that worked perfefctly.
There were a lot of codes which got tangled as there were around 18 dart files in this project put at the end we managed to order it perfectly and the project is all set to roll on its own.

Discussion