An Intelligent Crop Recommendation System Using RIC-CNN

Description:

This work proposes an intelligent crop recommendation system using a Residual Inception Convolutional Neural Network (RIC-CNN) to help farmers select suitable crops based on soil and environmental conditions. The system uses key parameters such as nitrogen (N), phosphorus (P), potassium (K), temperature, humidity, soil pH, and rainfall as input features.

The collected agricultural data is preprocessed by cleaning, scaling the features, and encoding crop labels. Feature scaling ensures uniform contribution of all inputs, while label encoding converts crop names into numerical values. The processed dataset is then split into training and testing sets.

The RIC-CNN model is designed to extract meaningful patterns from the input data using multi-scale convolution layers and residual connections, which improve learning efficiency and accuracy. The model is trained using the training data and evaluated using standard classification metrics to ensure reliable performance.

After training, the model, along with the scaler and label encoder, is deployed through a Flask-based web application. Users can log in, enter soil and climate details, and receive crop recommendations with confidence scores. The proposed system supports data-driven decision-making in agriculture and contributes to improved crop planning and productivity.

Input  : Soil and environmental parameters

Output  : Recommended crop name

Tags :

#soil, #environmental_parameters, #Nitrogen, #Phosphorus, #Potassium, #Temperature, #Humidity, #Soil_pH, #Rainfall, #soil_fertility, #climate, #weather_data, #agriculture, #farming, #crop_data, #machine_learning, #deep_learning, #RIC_CNN, #CNN, #model_training, #feature_scaling, #data_preprocessing, #classification, #prediction, #crop_recommendation, #confidence_score, #precision_agriculture, #smart_farming, #decision_support, #sustainable_agriculture, #yield_prediction, #web_application, #Flask, #real_time_system

References:

1.     Kumar, A., Sarkar, S., and Pradhan, C. (2019). Crop selection based on soil and environmental characteristics using machine learning techniques. International Journal of Advanced Computer Science and Applications, IEEE.

2.     Ramesh, D., and Vardhan, B. V. (2015). Analysis of crop yield prediction using data mining techniques. International Journal of Research in Engineering and Technology (IJRET).