Real-Time Noise Pollution Monitoring Using IoT, AI-Based Detection, and Camera Analytics

Abstract

Noise pollution has become a serious environmental problem in fast-growing cities and suburban areas. Long-term exposure to high noise levels is linked to stress, heart problems, issues with thinking, and a lower quality of life. As a result, it’s important to develop a system for continuous and intelligent monitoring. This research introduces as IoT-enabled real-time environmental noise monitoring system that combines acoustic sensing, smart alert generation based on thresholds, and camera-based location analysis. Ambient noise levels are collected using acousticsensorsconnected toaRaspberryPi,which arecalibrated against set limit values.When the system detects limit violations, it automatically generates alerts, records decibel level with timestamps and geocoordinates, and captures visual evidence to help identify regular noise sources. Lightweight machine learning classifier is included to tell apart short noise events from ongoing pollution,improvingdetection accuracy.Thesystems design focuses onlowpoweruseandmodularscalability,allowingittobedeployed in hospitals, schools, residential area, along highway, and at large public events. By combining intelligent sensing, adaptive classification,andvisualanalytics, thisresearch providesa practical framework for smart urban governance and effective noise pollution mitigation.

Introduction

Noise pollution has become a major environmental concern in rapidly growing urban and suburban areas, as prolonged exposure to excessive noise can cause stress, cardiovascular problems, cognitive impairment, and reduced quality of life. To address this issue, the study proposes an IoT-enabled real-time environmental noise monitoring system that integrates acoustic sensing, intelligent alert generation, and camera-based location analysis.

The system uses acoustic sensors connected to a Raspberry Pi OS-based platform to continuously measure ambient noise levels and compare them with predefined threshold limits. When noise levels exceed permissible limits, the system automatically generates alerts, records decibel values along with timestamps and geolocation data, and captures images to provide visual evidence of potential noise sources.

A lightweight machine learning classifier is incorporated to distinguish between temporary noise events and persistent noise pollution, thereby improving monitoring accuracy. The system is designed with low power consumption and modular scalability, making it suitable for deployment in hospitals, schools, residential areas, highways, and large public gatherings.

Conclusion

Thisresearchestablishesapracticalanddeployableframeworkforreal-time environmental noise surveillance by integrating IoT sensing, machine-learning-based classification, and camera-assisted spatial validation. Thesystemwasevaluated acrosstrafficcorridors, residential areas, and institutional environments using 1,020 recorded sound samples,comprisingboth backgroundnoise andviolationevents. After signal conditioning, the LM393 acoustic sensor demonstrated a measurement deviation of approximately ±3 dB, which is consistent with the expected performance of low-cost IoT sound sensors. Usingamplitude,frequencyvariation,andtemporalenergyfeatures,the trained classification model achieved an overall detection accuracy of 61.3%. It showed balanced precision and recall, indicating reliable baseline performance for detecting noise violations. About 45% of recordedeventsexceededthe70dBthreshold,confirmingthesystem’s ability to tell apart critical and non-critical acoustic conditions in real environments. The addition of camera-based spatial validation further improved the detection process by confirming more than half of the acoustically identified violations. This reduced uncertainty caused by transient or overlappingsoundsources.Despitethelimitations oflow-costsensors, using machine-learning-based filtering significantly improved performance compared to traditional threshold-only methods. With stable real-time operation and low alert latency, the proposed platform issuitableforongoing smartcitydeployment. Futureupdates,including adaptive calibration, larger datasets, and deep-learning models, can enhance system accuracy, robustness, and scalability.

References

[1] M.Anachkova,S.Domazetovska,Z.Petreski,andV.Gavriloski, “IoT-based urban noise monitoring using LoRaWAN technology,” Int. J. Smart City Environ. Monit., North Macedonia, Dec. 2020. [2] R.Maulidda,A.Abdurrahman,A.T.Wardhana,and M.A.Yahya, “Design and implementation of an IoT-based noise detection systemforlibraryenvironments,”J.InternetThingsSmartSyst., Indonesia, Feb. 27, 2024. [3] L.Choodarathnakara,S.Punitha,Vinutha,ShwethaB.D.,A.H. [4] J.Abhishek,andG.S.Sinchana,“IoT-basedreal-timemonitoring of air and noise pollution in urban areas,” in Proc. Int. Conf. Embedded Commun. Syst., India, 2022. [5] P.Meenalochini,“IoTandcloud-integratedsmartenvironmental monitoring system with predictive analytics,” J. Environ. Eng. Smart Technol., India, 2022. [6] T.Sepanosian,X. Küpers,P.Joza,F.Massah, andR.Bemthuis, “Real-time IoT-based emission monitoring architecture for construction sites,” in Proc. Sustainable Smart Infrastructure Conf., Netherlands, 2023. [7] Saddiwal, Z. Inamdar, M. Dhumal, O. Rajankar, and B. Shinde, “Design of a GPS-enabled smart noise monitoring system for urbanenvironments,” Int.J.Electron.Telecommun.Eng.,India, Apr. 2024. [8] R. M.Alsina-Pagès, R. Benocci, G. Brambilla, and G. Zambon, “Comparative evaluation of noise event detection algorithms in urban sound environments,” Appl.Acoust., vol. 181,Aug. 2021. [9] J.-M.Bozonnet,V.Hong,J.-F.Petiot,andP.Scalart,“Automatic sound recognition methods for intelligent urban noise monitoring,”IEEETrans.AudioSpeechLang.Process.,France, Aug. 2018. [10] P. Kumar, U. Pritam, R.Yadav, and G. Pandey, “NoiseTube:A participatorysensingplatformfornoisepollutionmappingusing mobile phones,” Int. J. Environ. Eng., India, 2014. [11] P.Sivaprakash,A.Purushothaman,M.S.Kumar, andJ.Prabhu, “IoT-based smart system for monitoring and control of air and noise pollution,” J. Smart Embedded Syst., India, 2023. [12] M.Alam,M. M.Islam, N.M.Nayan, andJ. Uddin,“Alow-cost IoT-based real-time environmental monitoring system for urban areas,” Sustain. Cities Soc., Oct. 2024. [13] K.Somu,M.Ananthi,R.Riventh,P.Sabapathi,A.Surya,andG. Thangabalu, “Intelligent GPS-based environmental monitoring using hybrid CNN–LSTM deep learning model,” Int. J. Artif. Intell. IoT, India, Jun. 2025. [14] J.RajagukgukandN.E.Sari,“Designofamicrocontroller-based soundnoiseleveldetectionsystem,”J.Phys.Instrum.,Indonesia, 2018. [15] G. Sirisha, V. Shravani, V. Rupavathi, V. S. Reddy, and V. M. Reddy,“SmartIoT-basedairandnoisepollutionmonitoringwith automaticpurification,”Int.J.SmartCityAppl.,India,Apr.2025. [16] R.Subraja,T.Meyyapan,A.Padmapriya,andS.SantoshKumar, “Design and implementation of an IoT-based noise pollution monitoring system,” J. Smart Environ. Monit., India, Oct. 2024.

Copyright

Copyright © 2026 Yatesh Ghadage, Dhanshri Kumbhar, Prathmesh Langde, Vaishnavi Midgule, Prof. Ashwini Deshmukh. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.