KissanKunj

Inspiration

As we have seen that RBI has shown a keen interest in including a digital currency-based Indian market. The given model satisfies the demands of the Central as well as state governments that are planning to make more decentralized marketplaces.

Recognizing the potential gains of Blockchain technology, The International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Eleven01 (India’s indigenous blockchain platform), and KHETHINEXT (e-agricultural service provider) have collaborated to employ blockchain technology to increase productivity and income of small-scale farmers in India.

This gave us the idea to create our own project with all of the features.

What it does

The solution that we came forward with was developing a system for selling the crops as NFT created as smart contracts in Solidity. Our plan of action is to create a marketplace where the farmer can directly sell his/her crops to the customer and track the crops as well. First, the User will Log-In(SAWO Authentication) to the NFT Marketplace and then You can Create your Own NFTs and sell them at a certain Price (MSP) Provided as taking account into the current trending analogy of that crop. And also You can Check if the crop is healthy or not and if it's not then you can also check, Which type of disease it has. Our solution also features Deep Learning-based Crop disease detection and Yield prediction models integrated with the WebApp. We also implemented a crop NewsAPI in our app using the news API. And After Someone buys that NFT, they Can then track the crop whole supply Chain.

How we built it

1)Starting from the frontend part of the app , we build it using next.js and initialized the boilerplate code using npx create-next-app .

2) Later we typed our own contracts using solidity language to build a marketplace of nfts.

3) Further we expanded our application by integrating contracts using web3 into our frontend app.

4)For running the app locally , running tests and debugging solidity code we configured hardhat . Hardhat Network is a local Ethereum network designed for development.

5)Further we used newsapi to fetch news related to agriculture to showcase in our application.

6)Then we further trained 2 models with very high accuracy for checking the quality of crops and predicting if there is any kind of disease. We hosted these models on heroku and integrated them with our application.

7)Moreover we were thinking of adding authentication in the application , thus we used SAWO Lab no password authentication for that .

Challenges we ran into

When we first started, we ran into a number of issues, including 1) Payments on the ETH Network were not being processed. 2) Difficulties with ML Model Deployment 3) There are a few issues with the Tracking System.

However, we were able to overcome all of our obstacles and make this project a huge success.

Accomplishments that we're proud of

We are happy with our achievements because we completed the entire project on time.

1) To our knowledge, this type of concept with this much capability has not been implemented anywhere.

2) The NFT MarketPlace was successfully launched.

3) The accuracy of all Machine Learning Models is a whopping 99.7 percent and are running completely fine.

4) The NewsAPI has been successfully implemented.

5) SAWO Authentication has been implemented successfully.

6) The tracking system is up and running.

Overall, we were able to overcome all of our obstacles and complete our Project as planned.

What we learned

During the hackathon, we learned a lot of new things, like how to incorporate machine learning models into web applications, how to construct our own DApp, how to create a tracking system for certain products, how to deploy on different test networks, and much more. We read the SAWO documentation, asked questions on the Discord channel, and then integrated it into our platform. This two-day hackathon has taught us far more than we could have learned in a month. In the middle, we became stuck in a variety of challenges, which ultimately improved our whole experience.

What's next for KissanKunj

We've considered deploying all of this web application in an App-Based Network and adding more machine learning models to it in order to check through a huge variety of crops and determine if they're healthy or not. We even intend to start with launching in the villages.