Smart City Gnosys

Smart city article details

Title	Towards Autonomous Road Maintenance: Cnn-Fl Integration For Pothole Detection Systems
ID_Doc	58038
Authors	Mehta S.; Singh A.
Year	2024
Published	2024 4th Asian Conference on Innovation in Technology, ASIANCON 2024
DOI	http://dx.doi.org/10.1109/ASIANCON62057.2024.10837744
Abstract	The motorists are mad at their cars, but the problem that they always face is the highway potholes. Therefore, the issue that necessitates the utilization of proper detection and repair procedures is the traffic jam it ignites in the city's streets, making it very congested. The suggested method of innovative use of Convolutional Neural Networks (CNN), combined with Federated Learning (FL), to find the optimal scenario for potholes is the one this research proposes. One of the finest features of the smart city idea is that the CNN + FL point about implementation can be scaled up without compromising users' privacy. More effective than the conventional method, this mechanism uses CNN, wherein not a single node trains the model, but the whole pool does. This leads to the final elimination of CNN-mon models. The comparison of the CNN+FL model and a basic CNN-based model resulted in about 2% on average er - the F1 score of 0.8562 demands that there should be a decent balance between the two models. A value of slightly more than 0.87 also reflects accuracy and recall. Deep neural networks can be supplemented with a federated learning system to improve the maintenance of urban infrastructure with concession. The existence of dissimilarities between countries and external factors shows the significance of further studies to put it at its finishing and best-applicable version. © 2024 IEEE.
Author Keywords	Data Privacy; FedAvg Approach; Federated Learning; Machine Learning; Pothole Detection; Privacy Preservation; Security

Similar Articles

Id	Similarity	Authors	Title	Published
16106	0.929	Lakshminarayanan S.; Konidhala J.	Convolutional Neural Network For Pothole Identification In Urban Roads	International Journal Of Advances In Signal And Image Sciences, 10, 1 (2024)
313	0.913	Hijji M.; Iqbal R.; Kumar Pandey A.; Doctor F.; Karyotis C.; Rajeh W.; Alshehri A.; Aradah F.	6G Connected Vehicle Framework To Support Intelligent Road Maintenance Using Deep Learning Data Fusion	IEEE Transactions on Intelligent Transportation Systems, 24, 7 (2023)
20646	0.903	Thamizharasi M.; Sethuraman R.; Sandhya A.	Distributed Framework For Pothole Detection And Monitoring Using Federated Learning: A Privacy-Preserving Edge Computing Approach	National Academy Science Letters (2025)
45593	0.897	Liu J.; Wang H.; Wei J.; Nan J.	Research On Road Pothole Recognition Based On Deep Convolutional Neural Networks	2024 IEEE 2nd International Conference on Image Processing and Computer Applications, ICIPCA 2024 (2024)
20645	0.891	Ezil Sam Leni A.; Shalen S.	Distributed Framework For Pothole Detection And Monitoring Using Federated Learning	2nd IEEE International Conference on Advances in Information Technology, ICAIT 2024 - Proceedings (2024)
204	0.89	Alshammari S.; Song S.	3Pod: Federated Learning-Based 3 Dimensional Pothole Detection For Smart Transportation	ISC2 2022 - 8th IEEE International Smart Cities Conference (2022)
44425	0.889	Bhosale S.B.; Ponnusamy S.	Real-Time Pothole Detection Using Yolov7: An Efficient Deep Learning Approach For Road Safety And Maintenance	2025 International Conference on Data Science and Business Systems, ICDSBS 2025 (2025)
44340	0.884	Mehajabin N.; Ma Z.; Wang Y.; Tohidypour H.R.; Nasiopoulos P.	Real-Time Deep Learning Based Road Deterioration Detection For Smart Cities	International Conference on Wireless and Mobile Computing, Networking and Communications, 2022-October (2022)
17906	0.882	Chu H.-H.; Saeed M.R.; Rashid J.; Mehmood M.T.; Ahmad I.; Iqbal R.S.; Ali G.	Deep Learning Method To Detect The Road Cracks And Potholes For Smart Cities	Computers, Materials and Continua, 75, 1 (2023)
37778	0.881	Obreja M.-E.; Dobrea D.-M.	Modified Resnet-50 For Training The Neural Network In Pothole Detection Using Deep Learning In Matlab	Proceedings of the 16th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2024 (2024)