India is the second largest agricultural producer and seventh largest exporter of agricultural goods. Agriculture plays a major role in the country’s economy as it provides employment to a lot of people, a major source of raw materials and is important for international trade. The share of agriculture in GDP increased to 19.9 per cent in 2020-21 from 17.8 per cent in 2019-20, the share of agriculture has increased to 20% after a gap of about 17 years.
Although the increase in GDP is appreciable, it cannot be deterred that many farmers commit suicides due to the less yield of crops. In February 2022, over 17,000 farmers between 2018 and 2020 committed suicide in different parts of the India. These numbers are highly alarming as farmers and farming is the backbone of a country like India.
In order to help the farmers with the types of crop that they can plant in their farm to increase the yield, and to also let them know of the kind of crops that can be grown along with the ones that are there for better yield we built various models using machine learning algorithms and the model with the best accuracy was chosen for the real time prediction.
Using deep learning models, convolution neural networks, disease of a plant is detected i.e; many times, farmers might not be able to detect the issue with their crop thus leading to the disease spreading to other crops and the entire yield produced being wasted. Therefore upon uploading the picture of the crop, the farmer will get complete information about the disease and how it can be mitigated by them. A user-friendly web application has been created using the Flask framework. This application will be available for the users at the comfort of their phone where they can know about what kind of crops to grow as well as how to mitigate the diseases acquired by the plants.
#Features
Django, python Mongodb machine learning deep learning android studio/ web development stack different sensors
Applied descriptive statistics visualization and predictive modeling has been applied to identify the crops that can be grown at a stipulated temperature and other natural conditions.
Since we are using transfer Learning there is being difficulties to reshape and resize the image in order to train the model
Faced problem to extract the weights of pre-trained model for alex-net architecture
It was quite difficult to get the real-time data as all the API's were paid. We were not able to efficiently put deep learning into use. Hence we have to go ahead with machine learning algorithms.
we are successfully able to predict the crop which are most suitable for the soil to be grown there and classify the plant diseases through leafs which will beneficial for the farmers.
We have made two models in our solution, in order to compare the accuracies. - In Deep CNN, the input data is passed into three convolution layers . The image is thereby optimised and the output is passed through relu activation layer. - In AlexNET, along with convolution layer, max pooling layer is used. The model with highest accuracy has been used.
fullstack development using FLASK and working with github and git, in front end we learn to build the website using html css and javascript
we will build a IOT device which will be present in farm and take the readings for the soil on real time manner and the same data will be analysed by ML and DL models which will help to give real time prediction as well
