Comparative performance evaluation of waste detection models in urban environments
DOI:
https://doi.org/10.24054/raaas.v15i1.2922Keywords:
Detection models, image processing, performance, accuracy, wasteAbstract
The objective of this paper was to evaluate and compare the performance of three debris detection models: Faster R-CNN ResNet101, SSD MobileNet V2, SSD MobileNet V2 FPNLite. An own set of images obtained from photographs taken in the Bogotá Channel in the city of Cúcuta was used to train the detection models. Key aspects such as accuracy, Recall, F1-Score, loss function, execution time and size of the models were evaluated. The model with the best performance characteristics was the Faster R-CNN ResNet101 model, with an F1-Score value of 83.8%, an accuracy of 89.3% and a Recall value of 78.9%.
Downloads
References
Aljabri, M., AlAmir, M., AlGhamdi, M., Abdel-Mottaleb, M., & Collado-Mesa, F. (2022). Towards a better understanding of annotation tools for medical imaging: a survey. Multimedia Tools and Applications, 81(18), 25877–25911. https://doi.org/https://doi.org/10.1007/s11042-022-12100-1 DOI: https://doi.org/10.1007/s11042-022-12100-1
Barba-Guamán, L. R. (2021). Uso de técnicas deep learning para reconocimiento de objetos en áreas rurales (Disertación Doctoral). Universidad Politécnica De Madrid. Madrid, España.
Bisong, E. (2019). Google Colaboratory. En Building Machine Learning and Deep Learning Models on Google Cloud Platform (pp. 59–64). Apress, Berkeley, CA. https://doi.org/https://doi.org/10.1007/978-1-4842-4470-8_7 DOI: https://doi.org/10.1007/978-1-4842-4470-8_7
Corso, C. L. (2009). Aplicación de algoritmos de clasificación supervisada usando Weka. Universidad Tecnológica Nacional, Facultad Regional Córdoba.
Daroca-Capell, T. (2014). Dimensión y características de los factores que inciden en la contaminación ambiental por residuos sólidos (Tesis de maestría). Universidad de San Martín de Porres. Lima, Perú.
El Tiempo. (2022). Remueven 4 toneladas de basuras del Canal de Bogotá. https://diarioeltiempo.co/nacional/remueven-4-toneladas-de-basuras-del-canal-bogota/1797/
Frutos, V., Pérez, B., Campos, S., Barrera, A., García, O., García, F., & Al-Kaff, A. (2023). Detección de residuos urbanos mal ubicados mediante visión por ordenador e IA. XLIV Jornadas de Automática, 855–860. https://doi.org/https://doi.org/10.17979/spudc.9788497498609.855 DOI: https://doi.org/10.17979/spudc.9788497498609.855
Hernández-Rodríguez, J. D. (2022). Mejora del modelo de detección de huecos y bolsas de basura basado en deep learning implementado en una Raspberry Pi (en Bogotá, Colombia) (Proyecto final de grado). Universidad de los Andes. Bogotá, Colombia.
Saiful Bahri, I. Z., Saon, S., Mahamad, A. K., Isa, K., Fadlilah, U., Ahmadon, M. A. Bin, & Yamaguchi, S. (2023). Interpretation of Bahasa Isyarat Malaysia (BIM) Using SSD-MobileNet-V2 FPNLite and COCO mAP. Information (Switzerland), 14(6). https://doi.org/https://doi.org/10.3390/info14060319 DOI: https://doi.org/10.3390/info14060319
Shamsuddin, M. F., Azami, M. H., Mohd Zaki, H. F., & Abdullah, N. A. (2022). Effect of Color Contrast to the Accuracy of SSD-MobileNetV2. International Journal of Intelligent Systems and Applications in Engineering, 10(3), 18–21.
Vrigazova, B. (2021). The Proportion for Splitting Data into Training and Test Set for the Bootstrap in Classification Problems. Business Systems Research, 12(1), 228–242. https://doi.org/https://doi.org/10.2478/bsrj-2021-0015 DOI: https://doi.org/10.2478/bsrj-2021-0015
Zhou, Y., Wen, S., Wang, D., Mu, J., & Richard, I. (2021). Object detection in autonomous driving scenarios based on an improved faster-RCNN. Applied Sciences (Switzerland), 11(24). https://doi.org/https://doi.org/10.3390/app112411630 DOI: https://doi.org/10.3390/app112411630
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 REVISTA AMBIENTAL AGUA, AIRE Y SUELO
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
