On the Use of Light Sensors and Pattern Recognition Techniques for Automated Detection of Street Lighting Lamps

Soares, Guilherme M.; Almeida, A. G. B.; Mendes, Ricardo Manuel Nogueira; Teixeira, Edmilson Costa; Braga, Henrique A. C.; Machado, Milena; Broetto, R. S.; Castro, M. V. H. B.; Filho, H. O. Gomes; Varejão, Flávio Miguel; Filho, José Gonçalves Pereira; Candeia, A. B.; Sousa, Ricardo J. Alves de

doi:10.1007/978-3-319-10948-0_4

Cited by 1 publication

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Part of the research presented in this work has been previously discussed in Soares et al [2][3][4], where the system's hardware and software are detailed, as well as in Broetto et al [9], which analyzed the application of image processing techniques to automatically detect street lighting luminaries, and Broetto and Varejão [10], which combines feature extraction and selection algorithms to optimize system classification accuracy.…”

Section: Related Workmentioning

confidence: 99%

“…Soares et al [4] used the same dataset, the same classifiers and the same hierarchical classifier architecture as in [3], but included a feature selection preprocessing step in order to select the best feature subset for improving the overall classification accuracy. They used stratified 10 fold cross-validation method for feature selection and the leave one out method for comparing the classifiers accuracy.…”

Section: Related Workmentioning

confidence: 99%

“…The system proposed in [2][3][4] was designed to enhance the efficiency and cost-effectiveness of the inspection process. Therefore, the hardware that constitutes the overall system consists basically of low-cost and off-the-shelf radiometric sensors and a camera capable of collecting various types of information about the lighting equipment under evaluation.…”

Section: Data Acquisition Equipmentmentioning

confidence: 99%

“…This device is designed to determine the type and wattage of lamps installed on poles used for public lighting. Soares et al [2][3][4] have described the main elements of this device, how it operates, and reported test results based on laboratory and field-collected data. These promising results suggest a rapid update of public lighting records, potentially reducing the commercial losses resulting from the lack of energy consumption information.…”

Section: Introductionmentioning

confidence: 99%

“…The collected data is processed by an intelligent algorithm that provides information about the location, type, and power of the lamps. Previous work [2][3][4] yielded relevant results but only a few algorithms and simpler validation methods were used in the experiments. The simpler validation methods may produce over optimistic results.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Analysis and Improvement of Machine Learning Models for Detecting Street Lighting Lamps

Silva,

Salgado,

Varejão

et al. 2023

Learn. Nonlin. Mod.

View full text Add to dashboard Cite

The Brazilian public lighting network is maintained by city halls. To bill the energy provided to city halls, energy distribution companies should maintain an updated database of network poles, their lamp types, and wattages. However, it is common to encounter issues with misinformation, where the company is not notified about changes in the public lighting network by city halls and cannot update its database appropriately. To mitigate commercial losses, companies have resorted to sending teams for manual infrastructure checks, which is an expensive, time-consuming, and unreliable process. In this regard, this work aims to optimize the models proposed in the literature capable of accurately classifying the type and wattage of lamps on public lighting poles based on data collected from radiometric sensors and a professional camera. Data is processed using traditional machine learning and deep learning algorithms, along with more sophisticated validation techniques such as data transformation and hyperparameter optimization to achieve improved results. Based on this methodology, the results demonstrate that models employing more robust algorithms (Support Vector Machine, XGBoost, Random Forest, and Multilayer Perceptron) can attain a final average accuracy of 80-86%. This confirms the usefulness of this methodology as an alternative solution to address the issue of public lighting billing.

show abstract