Get Free Synopsis PDF
We will prepare a detailed synopsis for your college submission. Includes problem statement, objectives, DFD, ER diagram, and more.
A mobile and web application that uses convolutional neural networks to identify crop diseases from leaf images in real-time. Farmers can photograph affected leaves and receive instant diagnosis with treatment recommendations, helping reduce crop losses.
Problem Statement
Indian farmers lose approximately 15-25% of their crop yield annually due to plant diseases that go undetected until significant damage has occurred. Most farmers lack access to agricultural experts for timely diagnosis. A technology solution that enables instant disease identification from smartphone photos can significantly reduce these losses and improve food security.
Objectives
Train a CNN model on a comprehensive plant disease dataset
Achieve at least 90% accuracy in disease classification
Develop a mobile-friendly interface for image capture and diagnosis
Build a treatment recommendation database
Create a history tracking system for farmers
Modules
Image Preprocessing
Resize, normalize, and augment leaf images for consistent model input.
Disease Classification Model
CNN model (ResNet50 or EfficientNet) trained on PlantVillage dataset for multi-class disease detection.
REST API
Flask/Django API endpoint that accepts images and returns predictions with confidence scores.
Mobile Interface
Camera integration for real-time capture, gallery upload, and result display with treatment info.
Treatment Database
Curated database of organic and chemical treatments mapped to each disease.
History & Analytics
Track past diagnoses, visualize disease patterns by region and season.
Dataset
PlantVillage Dataset (87,000+ images across 38 disease categories) available on Kaggle.
Expected Output
A working application where farmers can photograph diseased crop leaves, receive immediate identification of the disease with confidence percentage, and get actionable treatment recommendations. The system supports major crops including tomato, potato, corn, and rice with 38+ disease categories.
Future Scope
Integration with government agricultural portals, multilingual support for regional languages, drone-based aerial crop monitoring, IoT soil sensor integration, and a marketplace connecting farmers with pesticide suppliers.