The CropNosis System : for Precision Agriculture

The CropNosis System : for Precision Agriculture

Asst. Prof. Ashutosh *1, Khushi Srivastav*2, Priyanka*3, Aman Kumar*4

*2,3,4 Students, Department of Computer Science and Engineering , HMR Institute of Technology and Management, GGSIPU, Delhi, India.

*1 Assistant Professor, Department of of Computer Science and Engineering, HMR Institute of Technology & Management, GGSIPU, Delhi, India.

I.                                           ABSTRACT

The global agricultural sector is currently undergoing a multitude of challenges that are extreme in terms of severity, and they include climate change, soil degradation, and a rising risk from plant disease, among others. These challenges collectively threaten the future of food and the economy. In this context, we need CopNosis, which is framework powered by artificial intelligence through a web application that allows farmers to rely on data in making decisions that would lead to increased crop productivity and sustainable farming practice.

The CropNosis system comprises of the following three components:

(1)                                              Crop Recommendation which suggests the most suitable crop by considering the changing (NPK, pH, temperature, humidity, rainfall) environmental and soil parameters,

(2)                                                Fertilizer Guidance which gives the top customized fertilizers for the given soil and crop conditions, and

(3)                                               Plant Disease Detection which uses deep learning based on Convolutional Neural Networks (CNN) to process a picture of the leaf and predict whether the disease will cause failure of the leaf or not.

Support is offered through the CropNosis system to different machine learning (ML) models on a variety of datasets including those collected from

Kaggle and Plant Village for the purpose of achieving a high level of accuracy in its recommendations and detections. The primary aim of this research is to show just how much an integrated ML framework can be of assistance in the transformation of traditional farming into more efficient, less costly and, above all, more resilient precision agriculture practice which is expensive in the long run.

KEYWORDS

Precision Agriculture, Machine Learning, Deep Learning, Convolutional Neural Networks (CNN), Crop Recommendation, Fertilizer Guidance, Plant Disease Detection, Random Forest, Decision Tree, CropNosis, Kaggle, Plant Village.