This project is an Arduino-based Emotion Detection System that estimates a person's emotional state in real-time by analyzing:
- 🫀 Pulse Rate
- 🌡️ Body Temperature
- ⚡ (Optional) GSR (Galvanic Skin Response)
It classifies emotions such as Happy, Sad, Relaxed, and Angry based on physiological data. Ideal for prototypes in health tech, affective computing, or wearable devices.
- Real-time emotion classification using body vitals
- Detects and logs:
- Heart Rate (BPM)
- Body Temperature (°C)
- GSR Value (Optional)
- Pulse Wave Signal
- Emotion mapping based on BPM
- Serial output in CSV format for easy logging
- Pulse waveform filtering via custom digital filter
| Component | Purpose | Connection |
|---|---|---|
| Arduino UNO | Microcontroller | — |
| Pulse Sensor | Heart rate detection | A0 |
| One Wire Temp Sensor | Body temperature measurement | D5 (Digital) |
| LED | Blinks on heartbeat | D13 |
| GSR Sensor | Measures Skin Conductivity | A1 |
| Breadboard | Mounting components | — |
| Jumper Wires | Connections between components | — |
| USB Port Cable | Upload code & power the board | USB port |
| Heart Rate (BPM) | Emotion |
|---|---|
| 85 – 125 | 😄 Happy |
| 80 – 90 | 😢 Sad |
| 65 – 79 | 😌 Relaxed |
| Above 125 | 😡 Angry |
Threshold values can be calibrated for individual differences.
- Connect components on breadboard as per pin table
- Use 4.7kΩ pull-up resistor with DS18B20 between D5 and VCC
- GSR sensor connected to A1
- Pulse Sensor to A0
- LED to D13
Install the following libraries in Arduino IDE:
- PulseSensor Playground
- OneWire
- DallasTemperature
- Upload the
.inocode - Open Serial Monitor at 9600 baud
- View real-time emotion prediction and vitals
- Emotion logic is based on heart rate only (can be extended to use GSR + temperature)
LABEL,Computer Time,Heart Rate,Body Temperature,GSR,Pulse
DATA, TIME,78,36.7,520,0.42,RELAXED