Efficient Edge Deployment of Quantized YOLOv4-Tiny for Aerial Emergency Object Detection on Raspberry Pi 5

TL;DR

This paper demonstrates that a quantized YOLOv4-Tiny model can be effectively deployed on Raspberry Pi 5 for real-time aerial emergency object detection, balancing speed, power, and accuracy.

Contribution

It introduces a quantized YOLOv4-Tiny model optimized for edge deployment on Raspberry Pi 5, enabling real-time emergency detection with low power consumption.

Findings

Inference time of 28.2 ms per image

Average power consumption of 13.85 W

Robust detection accuracy across emergency classes

Abstract

This paper presents the deployment and performance evaluation of a quantized YOLOv4-Tiny model for real-time object detection in aerial emergency imagery on a resource-constrained edge device the Raspberry Pi 5. The YOLOv4-Tiny model was quantized to INT8 precision using TensorFlow Lite post-training quantization techniques and evaluated for detection speed, power consumption, and thermal feasibility under embedded deployment conditions. The quantized model achieved an inference time of 28.2 ms per image with an average power consumption of 13.85 W, demonstrating a significant reduction in power usage compared to its FP32 counterpart. Detection accuracy remained robust across key emergency classes such as Ambulance, Police, Fire Engine, and Car Crash. These results highlight the potential of low-power embedded AI systems for real-time deployment in safety-critical emergency response applications.