International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 13 Issue: 08 | Aug 2026
p-ISSN: 2395-0072
www.irjet.net
Urban Shield: An IoT-Based Intelligent Manhole Detection and Monitoring System NAGASHREE K S¹, NIDHI², NISHITHA P P³, Dr. THRISHA V S⁴ ¹²³ UG Student, Department of Computer Science and Engineering, Sir M Visvesvaraya Institute of Technology, Bengaluru, Karnataka, India ⁴ Assistant Professor, Department of Computer Science and Engineering, Sir M Visvesvaraya Institute of Technology, Bengaluru, Karnataka, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract-Manhole covers are essential components of urban
on roads and pedestrian areas, creating risks for vehicles and pedestrians. Problems within underground drainage systems can also create unsafe conditions for sanitation workers due to flooding, water accumulation, and exposure to harmful gases. Existing research has identified these issues as important challenges in the management and maintenance of urban manhole infrastructure.
infrastructure, and their absence, displacement, or damage can create serious safety risks for pedestrians, vehicles, and sanitation workers. Conventional inspection methods rely mainly on periodic manual checks, making continuous monitoring difficult across large urban areas. Existing IoTbased approaches improve monitoring by collecting parameters such as cover status, water level, flooding, gas concentration, and movement through sensors. However, most of these systems primarily depend on sensor measurements and threshold-based alerts, with limited integration of realtime visual analysis and predictive capabilities. This paper presents Urban Shield, an IoT-based intelligent manhole detection and monitoring system that combines IoT sensing, real-time image acquisition, computer vision, artificial intelligence, and a web-based monitoring platform. The system uses IoT devices to collect field information, Python for AI and computer-vision processing, and a trained YOLO model to detect and analyze manhole conditions from captured images. The processed information is integrated with backend and database services and presented through a MERN-based realtime dashboard. The proposed framework also incorporates AI-based analysis for supporting prediction or risk assessment using collected sensor information and detected visual characteristics. By combining physical sensing with visual intelligence, Urban Shield aims to reduce dependence on manual inspection and provide a more integrated approach to real-time urban infrastructure monitoring. Experimental performance measures are to be obtained through model and system evaluation and are not reported without supporting experimental evidence.
Conventional assessment largely relies on scheduled physical inspections and reports from individuals, making continuous observation difficult across widely distributed manholes. This approach becomes difficult when a city contains a large number of manholes distributed across different locations. Such an approach requires substantial human effort and cannot provide uninterrupted observation of every monitored location. Earlier research has therefore focused on developing automated monitoring systems that can detect abnormal conditions and provide timely information to responsible authorities. IoT has increasingly been investigated as a means of automating monitoring activities within urban infrastructure. These systems employ sensors to observe parameters including cover movement, inclination, flooding, water level, gas concentration, and open or missing covers. The collected information can be transmitted through communication technologies such as Wi-Fi, GSM, LoRa, and LoRaWAN to monitoring platforms or management authorities. For example, one of the reviewed systems uses acceleration, flooding, and positioning information and transfers the collected status through LoRa to an IoT cloud platform, where abnormal conditions can be monitored and reported.
Key Words: IoT, Manhole Detection, Computer Vision, YOLO, Deep Learning, Smart City, Real-Time Monitoring, Predictive Analysis, Urban Infrastructure, Intelligent Monitoring.
Although sensor-based monitoring provides useful information about the physical condition of a manhole, sensor readings alone may not describe the complete visual situation at the monitored location. Camera-based image acquisition can provide additional information about the surrounding environment and the visible condition of a manhole. The reviewed literature also includes computervision and deep-learning approaches for automated manhole identification, showing the potential of visual intelligence for infrastructure monitoring.
1. INTRODUCTION Manholes are important components of urban infrastructure because they provide access to underground drainage, sewage, communication, electrical, and other utility networks. Despite their limited physical size, manholes can have a significant impact on public safety and urban infrastructure operations. Missing, displaced, damaged, or improperly positioned manhole covers can expose openings
© 2026, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 420