CamelGuard: AI-Powered Camel Detection System for Road Safety Using YOLO and IoT Sensors

Document Type

Conference Proceeding

Source of Publication

2026 17th Student Research Conference on Applied Computing Src 2026

Publication Date

4-22-2026

Abstract

The United Arab Emirates (UAE) faces critical road-safety risks from camel-vehicle collisions on its desert highways, with camels accounting for 83% of all animal-related road accidents in the country. Existing passive measures, such as roadside fencing and warning signs, have proven insufficient, particularly in remote areas and at night. This paper presents CamelGuard AI, an autonomous, offline-capable, edge-computing detection and driver alert system. The system integrates a YOLOv11-Tiny deep learning model, an HC-SR04 ultrasonic proximity sensor, a Raspberry Pi NoIR camera with 940nm infrared LED illumination, a DHT11 environmental sensor, and a Flask web monitoring dashboard, all running locally on a Raspberry Pi 5 within a custom 3D-printed PLA+ enclosure. Two specialized models (day.pt and night.pt) were trained on a Roboflow-managed camel dataset of 230 images with 2,040 labeled instances, achieving [email protected] of 0.869 and 0.842, respectively. Prototype testing confirmed 94.7% daytime and 91.0% nighttime detection accuracy with sub-200 ms alert latency. The complete prototype was assembled for 834 AED (~$227 USD), representing more than 180× the cost reduction compared to commercial radar alternatives. Stakeholder validation from Sharjah Police further confirmed operational feasibility and national relevance.

ISBN

[9798319510167]

Publisher

IEEE

Disciplines

Computer Sciences

Keywords

camel detection, edge AI, infrared night vision, IoT, object detection, Raspberry Pi, road safety, sensor fusion, YOLOv11

Scopus ID

105043147913

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS