<![CDATA[The increasing volume of household waste that is not managed effectively in Indonesia poses a serious environmental problem, primarily due to the low public awareness in sorting waste by type. Conventional Waste Banks face operational constraints such as manual recording processes and inefficient waste collection. This research aims to develop an integrated solution in the form of a mobile application that combines digital Waste Bank management features with Computer Vision (CV) technology for waste classification. The application is designed to facilitate users in submitting recyclable waste collection requests and provides an educational feature for automatic waste type recognition via the smartphone camera. The development method includes the design of an integrated information system (user application, web admin, and database) and the implementation of a Deep Learning model (CNN) optimized using TensorFlow Lite to run in real-time on Android devices. The system functionality test results indicate that all main features (registration, transactions, balance mutation, and collection requests) operate normally according to the design specifications. This research contributes to providing a functional, transparent system prototype that can serve as an effective educational tool to encourage community participation in sorting and managing waste at the source.]]>
Copyrights © 2025
