Design and Implementation of an AI Emotion Recognition-Based Smart Diffuser System for Vehicles

This paper proposes an AI emotion recognition-based smart diffuser system for vehicles that monitors a driver's emotional and drowsiness states in real time, dynamically controlling fragrance diffusion, visual feedback, and auditory alerts. Traditional vehicle diffusers are typically limited to manual operation or fixed periodic spraying, making it difficult to actively adapt to changing driver conditions. To overcome these limitations, the proposed system integrates a Raspberry Pi-based AI recognition unit, an ESP32-based hardware control unit, a Flask REST API server, and a React Native mobile application. The Raspberry Pi captures camera frames to detect the driver’s face and analyze emotions and drowsiness, while the ESP32 manages RGB LEDs, fragrance diffusion modules, a cooling fan, a buzzer, and an OLED display via UART commands. Experimental results demonstrate that the emotion classification model achieves an accuracy of 83.46%, and the drowsiness detection module achieves over 90% accuracy. These findings validate that the proposed system serves as an effective, intelligent in-vehicle environment control device capable of actively responding to the driver's physiological and emotional states.

emotion recognition, smart diffuser, ESP32, drowsiness detection

1. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, "MobileNetV2: Inverted Residuals and Linear Bottlenecks," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4510-4520, 2018. [Google Scholar] [Crossref]

2. S. Li and W. Deng, "Reliable Crowdsourcing and Deep Locality-Preserving Learning for Facial Expression Recognition," IEEE Transactions on Image Processing, vol. 28, no. 1, pp. 356-370, 2019. [Google Scholar] [Crossref]

3. A. Mollahosseini, B. Hasani, and M. H. Mahoor, "AffectNet: A Database for Facial Expression, Valence, and Arousal Computing in the Wild," IEEE Transactions on Affective Computing, vol. 10, no. 1, pp. 18-31, 2019. [Google Scholar] [Crossref]

4. P. Viola and M. Jones, "Rapid Object Detection using a Boosted Cascade of Simple Features," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 511-518, 2001. [Google Scholar] [Crossref]

5. TensorFlow, "TensorFlow Lite Guide," Google Developers. [Google Scholar] [Crossref]

6. OpenCV, "Open Source Computer Vision Library Documentation." [Google Scholar] [Crossref]

7. Espressif Systems, "ESP32 Technical Reference Manual." [Google Scholar] [Crossref]

8. Flask Documentation, "Flask Web Development Framework." [Google Scholar] [Crossref]

9. React Native Documentation, "Learn Once, Write Anywhere." [Google Scholar] [Crossref]

10. Raspberry Pi Foundation, "Raspberry Pi Documentation." [Google Scholar] [Crossref]

• What the Desert Fathers Teach Data Scientists: Ancient Ascetic Principles for Ethical Machine-Learning Practice
• Comparative Analysis of Some Machine Learning Algorithms for the Classification of Ransomware
• Comparative Performance Analysis of Some Priority Queue Variants in Dijkstra’s Algorithm
• Transfer Learning in Detecting E-Assessment Malpractice from a Proctored Video Recordings.
• Dual-Modal Detection of Parkinson’s Disease: A Clinical Framework and Deep Learning Approach Using NeuroParkNet