The problem FarmX solves

In India, we all know that Agriculture is the backbone of the country and it is said that the more the farmers of a country prosper, the more the country prospers. We understand that the farmers of India need better and more efficient tools to help them increase their crop yield and produce better quality crops. Machine learning tools can be employed to help farmers choose which crops to grow wisely and help them judge the quality of their products so that they may get the best price for their produce.

Ideas/Models that solve the problem related to it:

1. Helping farmers irrigate the farm by analyzing the optimum moisture over a week using machine learning and image processing.
2. To alert the farmers about the weather condition and whether it is suitable for the growth of a particular crop or not, with the help of the website that we built. It also provides information about different government provisions and schemes for farmers.
3. Precision Farming and Product Analysis: A tool that helps farmers employ better strategies and provide them proper guidance about crop rotation, water management, etc.

Challenges we ran into

As our project was hardware-based we faced a lot of issues regarding calibrating and debugging the code.

*Making a web interface with constantly changing colors according to the quality of crops across the whole field was a difficult task as we got many errors in between but we finally solved it by creating a different CSV for each plant(section of the plant)

*Simultaneous serial printing and serial writing on the same Arduino was a bit difficult and we had to drop the plan of using an led to represent the crop matrix(3x3) we had in mind, instead we displayed the motor state in the website.

*To cut down the costs, we had to use a single sensor to record multiple sectors of the crop and that made it difficult to code, to solve that problem we considered a section at a time.

*As our project calculates the optimum moisture for the crop based on grading and statistics, it was difficult for us to get a wide array of values for every temperature and humidity level.

*The local atmospheric conditions kept changing and it was difficult to calculate the optimum moisture and simultaneously compare it with the current(local) moisture.