An Arduino Nano collects sensor data and transmits it to a Raspberry Pi 4, which serves as the central processing unit. The Raspberry Pi performs image processing and machine learning–based classification to determine the freshness status of fruits (good or spoiled). The camera enables live video streaming, allowing users to visually inspect the fruits remotely.

All collected data, including environmental parameters, gas concentration, and fruit condition, are displayed on a web-based dashboard in real time. The system also includes an LCD for local display and a buzzer alert to notify users when spoilage is detected. By combining sensor fusion and AI-based image analysis, the proposed system enhances food safety, reduces waste, and enables smart refrigerator management through remote monitoring and decision-making.

The main objectives of this project are:

• To design and develop an intelligent refrigerator monitoring system for detecting spoilage in fruits like apples and bananas.
• To integrate multiple sensors (gas, temperature, humidity, IR, and NIR) for real-time environmental and freshness monitoring.
• To implement AI/ML-based image processing techniques for classifying fruits as fresh (good) or spoiled (bad).
• To enable live video streaming of fruits using a camera for remote visual inspection.
• To develop a web-based dashboard for displaying sensor data, fruit selection, and spoilage status in real time.
• To provide early warning alerts (buzzer/notification) when spoilage or abnormal conditions are detected.
• To ensure seamless communication between Arduino Nano and Raspberry Pi for efficient data processing.
• To reduce food wastage by enabling timely detection and monitoring of fruit quality.
• To create a low-cost, user-friendly smart refrigeration system for domestic and commercial use.

The major building blocks of this project are:

1. Regulated power supply.
2. AC – DC converter.
3. Raspberry pi4.
4. SD card.
5. Pi camera.
6. Buzzer.
7. Arduino Nano.
8. MQ-135 sensor.
9. Methane sensor.
10. DHT11 sensor.
11. IR sensor.
12. NIR sensor.
13. Arduino Nano.
14. LCD display.
15. Buzzer.

Software’s used:

• Raspbian OS.

• Machine learning & Artificial Intelligence (AI) algorithm.
• Express SCH for Circuit design.
• WEB application.

Block Diagram:

video:

The DHT-11 sensor is used to measure temperature and humidity levels, which can help in detecting fire or flooding risks. The MQ135 sensor detects harmful gases, such as carbon monoxide and ammonia, which can be indicators of chemical leaks or environmental hazards. The vibration sensor helps in detecting seismic activity or tremors, which are critical in earthquake monitoring. The water level sensor detects changes in water levels, which is useful in flood-prone areas.

All sensor data is collected and processed by the Arduino Nano, which transmits the data to a cloud platform through the ESP8266 Wi-Fi module. The real-time data is displayed on an LCD screen for immediate monitoring. This disaster monitoring system provides early warnings for various types of disasters, enabling quicker responses to mitigate damage. The use of IoT ensures that data can be accessed remotely, allowing for continuous surveillance and prompt disaster management.

This system aims to improve disaster preparedness by providing accurate, real-time information that can aid in timely decision-making, contributing to enhanced safety and risk management in disaster-prone areas.

The major building blocks of this project are:

• Regulated power supply
• ARDUINO nano Microcontroller.
• DHT-11
• MQ135
• ESP8266
• VIBRATION SENSOR
• WATER LEVEL SENSOR
• LCD display.
• LED Indicators.
• Reset button.

Software’s used:

• Embedded C programming.
• Arduino nano for dumping code into Micro controller.
• Express SCH for Circuit design.

Block diagram: