Recycling contamination presents a significant challenge by increasing operational costs and diminishing the efficiency of material recovery facilities (MRFs). This proposed research aims to address the pressing need to enhance the quality of recyclable materials and improve recycling efficiency in Florida. The central objective is to develop a machine learning–based image recognition tool designed to identify contaminants within recycling streams and lithium batteries, and explore the best messaging for recycling encouragement. The tool will be constructed using a lightweight convolutional neural network (CNN), specifically optimized for integration into a mobile application for use by residents and transfer station operators during the collection and sorting process. By incorporating explainable artificial intelligence (AI) techniques, such as heatmap visualizations, the application will offer intuitive feedback to users, thereby fostering reductions in contamination at the household level. This study builds upon prior efforts involving both laboratory-generated recyclable images and real-world cart-tagging data, adapting the CNN model for mobile deployment and validating its performance through user testing across Florida. Additionally, the proposed work aims to continue exploring the best way to communicate with the residents for recycling encouragement. The project is structured around four key tasks: (1) image database development from the curbside recycling bins and transfer stations for model training; (2) image model selection, training, and optimization for contamination identification; (3) image tool testing, user feedback integration, and field validation; and (4) motivational messaging strategies for recycling engagement in the communities. By enhancing contamination and lithium batteries detection and public participation, the project aims to advance recycling outcomes and provide actionable data for policymakers, contributing to broader research agendas in artificial intelligence applications and sustainable waste management systems.