Automated Waste Classification with Tensorflow and Mobilenet: A Step Towards Sustainable Development Goal 12

Waste mismanagement is a critical environmental concern, with manual sorting often inefficientand prone to error. This paper presents an automated waste classification system usingConvolutional Neural Networks (CNNs) and transfer learning to automatically categorize wasteinto classes such as plastic, glass, organic, metal, and paper. The model was trained on a combineddataset sourced from the TrashNet repository and locally collected images, with data augmentationapplied to improve robustness under varying conditions. Experimental results demonstrate highclassification accuracy, outperforming traditional machine learning baselines. The system wasfurther optimized for deployment on resource-constrained devices, enabling real-time operationon platforms such as Raspberry Pi. By improving waste segregation at the source, this approachsupports Sustainable Development Goal 12 (Responsible Consumption and Production) andpromotes circular economy practices. Future work will focus on expanding classificationcategories and integrating IoT-enabled smart bin technology.