System Design
The platform is organised into connected layers for data collection, processing, intelligent analysis and presentation.
Data Acquisition
Sensors collect soil moisture, temperature and humidity. Users may also upload plant images.
Data Processing
Inputs are validated, cleaned and transformed before analysis.
Intelligence Layer
Machine-learning models predict irrigation and classify plant health.
Application Layer
The dashboard displays readings, alerts, predictions and recommendations.
Data Flow
Environmental readings and images enter the system, pass through validation and pre-processing, and are then analysed by the relevant model. Results are stored and displayed through the website. The gardener remains responsible for the final watering or treatment decision.
System Modules
- Sensor-data module
- Data-processing module
- Irrigation-prediction module
- Plant-image analysis module
- Web-interface module
Database Structure
The database stores sensor ID, timestamp, environmental readings, irrigation recommendations, prediction confidence and plant-health results. Optional user or garden records may be added later.
User Interface
The website uses consistent navigation, readable text, clear headings and responsive layouts. It presents information in cards, tables and image panels so that non-technical users can understand the results.
Reliability and Security
Validation checks help identify unrealistic readings. A deployed version should add authentication, secure communication and protected storage while collecting only necessary information.