

What if a farmer 👨🌾could identify crop pests, receive fertilizer recommendations, and estimate crop yield🌾—all from a single AI-powered platform?
Farmers often face critical challenges in identifying crop diseases and pests, selecting the right fertilizer based on soil conditions, and predicting crop yield before harvest. Making the wrong decision at any stage can affect productivity, increase costs, and reduce profits.
To address these challenges, we developed
AgroVision🌱AI – Smart Agriculture🌾: AI-Driven Pest Prediction, a web-based application that combines Artificial Intelligence and Machine Learning to support smarter farming decisions.
Key Features
▫️AI-Based Pest Detection🐞
Upload a crop/pest image.A CNN model based on MobileNetV2 classifies the pest.
Displays the detected pest, confidence score, severity level, and recommended pesticides.
▫️Smart Fertilizer Recommendation👨🌾
Takes crop type, soil type, temperature, humidity, moisture, and NPK values as input.
Recommends the most suitable fertilizer to improve crop health and productivity.
▫️Crop Yield Prediction🌾
Predicts expected crop yield using agricultural parameters such as crop type, season, rainfall, area, fertilizer usage, and pesticide usage with a machine learning regression model.
▫️Location-Based Weather Updates🌦️
Displays the current weather conditions based on the farmer's location, helping farmers plan irrigation, fertilizer application, and other agricultural activities more effectively.
The attached video demonstrates the complete workflow—from user registration and login to dashboard navigation, AI-based pest detection, fertilizer recommendation, and crop yield prediction.
#MachineLearning #ArtificialIntelligence #SmartAgriculture #AgriTech #ReactJS #Flask #TensorFlow #Keras #ScikitLearn #Python #WebDevelopment #FinalYearProject
7 Août 2026 à 17h15
Voir sur LinkedIn →