Overview
Anar X is an AI-driven agriculture platform developed to help pomegranate farmers make informed decisions throughout the crop lifecycle. The platform combines computer vision, machine learning, and real-time agricultural data to provide intelligent disease diagnosis alongside current market insights.
At the core of Anar X is a custom-trained deep learning model built using a curated dataset of real pomegranate plant images. The dataset includes healthy crops along with multiple disease classes, enabling the model to recognize visual symptoms and accurately classify plant diseases from smartphone images.
Once an image is captured or uploaded, the application preprocesses it, performs inference using the trained model, predicts the disease with a confidence score, and presents the farmer with detailed information about the disease, recommended treatments, and preventive measures.
To make the platform more practical beyond disease diagnosis, Anar X also integrates official government agricultural market APIs to retrieve real-time pomegranate market prices from different mandis. Farmers can monitor current market rates, compare prices across locations, and make more informed decisions about when and where to sell their produce.
By combining AI-powered crop health analysis with live market intelligence, Anar X provides a single platform that supports both cultivation and post-harvest decision-making.
Key Features
- Custom-trained deep learning model for pomegranate disease classification
- Disease detection using real crop images
- Image preprocessing and AI inference pipeline
- Prediction confidence score
- Disease information and treatment recommendations
- Preventive farming guidance
- Real-time pomegranate market prices from official Government APIs
- Market comparison across agricultural mandis
- Farmer-friendly mobile interface
- Regional language support
Technologies Used
Development Process
Data Collection
Collected and organized real-world pomegranate disease images.
Data Preparation
Cleaned, labeled, and augmented the dataset.
Model Training
Trained and evaluated a convolutional neural network for disease classification.
Optimization
Optimized the model for accurate predictions.
Integration
Integrated the trained model into a Flutter application.
Market Data
Connected official government agricultural market APIs to fetch live mandi prices.
UI/UX Design
Designed an intuitive mobile experience focused on accessibility and ease of use for farmers.
What I Learned
"Anar X allowed me to experience the complete lifecycle of building an AI-powered product, from preparing datasets and training deep learning models to integrating external government APIs, deploying intelligent inference pipelines, and designing a production-oriented mobile application. The project reinforced the importance of combining machine learning with reliable real-world data to build software that delivers practical value to its users."