Advanced Single-View Image-Based Framework for Volume Estimation in Urban Solid Waste Management
Julio Leite Azancort Neto, Romário da Costa Silva, Thalita Ayass de Souza, Carlos André de Mattos Teixeira, Evelin Helena Silva Cardoso, Jasmine Araújo, Carlos Renato Lisboa Francês · 2024
Efficient solid waste management is crucial for making the city a clean and sustainable environment. This paper presents a methodology composed of well-established algorithms for volume estimation in urban solid waste management using single-view images. The proposed system was built using state-of-the-art model-based algorithms including instance segmentation, depth estimation, and point cloud-based volume calculation The methodology demonstrates the ability to accurately estimate the volume of individual and multiple plastic bags containing municipal solid waste. We evaluated our approach using real-world data. Numerical results showed that the proposed system is promising even in complex scenarios. Despite challenges, such as: manual distance rescaling and limited datasets, our system holds considerable potential for further refinement and enhancement targeting scenarios as complex as real urban environments. The proposed methodology contributes to advancing management technologies in smart cities.