Citation

Azmawi, Azib Jazman and Mohd-Isa, Wan Noorshahida and Abdul Rahman, Abdul Aziz (2025) Safety Compliance Assessment of Fibre Distribution Panel Installations Using Computer Vision. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 468.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The telecommunication industry in Malaysia plays a pivotal role in providing essential Internet and communication services. Internet Service Providers (ISPs) are increasingly adopting on Fibre Distribution Panels (FDPs) to deliver high-speed Internet to homes and businesses. Each new FDP installation requires an assessment to ensure safety and readiness of these setups. However, current methods rely on manual measurements, making the process inefficient and resource-intensive. While automated safety assessment systems are common in other industries, their application in the telecom sector remains largely unexplored. This study addresses the hap by introducing a computer vision-based approach to automate the safety distance assessment for FDP installations. With deep learning models, the system detects components and applies low-level image processing, the Line-Line intersection algorithm and Euclidean Distance to extract object width in pixel units. The system then checks for violations by determining whether other components violates the minimum required spacing set by the company. Experimental results using real-world data demonstrate that the proposed method achieves 0.6 accuracy in component width retrieval at almost 12s per installation. Our overall installation quality assessment error rate is less than 0.35, mostly owing to object detection errors in identifying essential components. The system also cuts assessments time in half compared to traditional methods. This research enhances efficiency safety and service quality in the telecom industry, offering a scalable solution for reliable and timely FDP installation inspections.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	automatic assessment, distance measurement, computer vision, telecommunication, Line-Line Intersection
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	19 Mar 2026 00:39
Last Modified:	19 Mar 2026 00:39
URII:	http://shdl.mmu.edu.my/id/eprint/15502

Downloads

Edit (login required)