Computer Vision System Based on OpenCV for Solid Waste Classification in Piura, Peru

Garcia-Olivares M.; Carmen-Livia Y. · Journal of Advances in Information Technology (2026) · DOI: 10.12720/jait.17.9.1676-1690

Resumen

Inadequate solid waste management poses a threat to public health and environmental sustainability in urban contexts. This study aimed to determine how a computer vision system based on Open Source Computer Vision Library (OpenCV) reduces processing time and improves accuracy in solid waste classification in Piura, Peru. An applied study with a quantitative approach was conducted with operators. Data were collected using observation guides and a validated questionnaire. The results showed significant improvements: the number of waste items classified per minute increased by 80%, and the average processing time per batch was reduced by 46.8%. The system achieved an overall accuracy of 83.2%, with classification accuracy rates exceeding 70% in all evaluated categories. It is concluded that computer vision based on OpenCV effectively optimizes both speed and accuracy in solid waste classification, thereby significantly improving urban environmental management.

Instituciones: Universidad César Vallejo