Deep Learning based Cotton Plant Pest Classification and fertilizer type prediction system

This project presents a system designed for cotton plant pest classification and fertilizer type prediction using deep learning and machine learning techniques. The pest classification module uses convolutional neural networks (CNN) and DenseNet architecture to analyze images and identify pest types, including aphids, armyworm, beetle, bollworm, grasshopper, mites, mosquito, sawfly, and stem borer. The fertilizer prediction module utilizes environmental and soil-related parameters such as temperature, humidity, moisture, soil type, crop type, nitrogen, potassium, and phosphorous to classify the suitable fertilizer type. The system is implemented using Python with Flask for backend development, and HTML, CSS, and JavaScript for frontend development. A user interface is developed with modules for registration, login, pest classification, fertilizer prediction, and logout functionality. The system aims to provide accurate pest classification from input images and fertilizer label prediction based on tabular input. This approach reduces the dependency on manual identification methods and improves efficiency in managing crop conditions through automated classification.

Keywords: Deep Learning, Pest Classification, CNN, DenseNet, Fertilizer Prediction, Soil Parameters, Cotton Crop, Flask, Image Processing, Classification System.