Table 1 Trash classification in existing literature.

RecycleNet¹⁷

Deep learning architecture tailored for trash classification using TrashNet dataset

Achieved high accuracy in classifying recyclable objects

Improved trash classification in waste management

Hybrid approaches^18,19

Investigation of Multilayer Hybrid Systems and CNNs

Effectiveness of hybrid models in trash classification

Enhanced understanding of image features’ impact

ResNeXt models²⁰

Focus on medical waste classification

High accuracy in identifying medical waste types

Improved management of medical waste

DSCR-Net^20,21

Utilization of open-source datasets and single-object identification

Development of open-source datasets and identification techniques

Enhanced accessibility and efficiency in trash classification

GCNet²², DSCAM²³

Deep learning-based trash classification with high accuracy rates

Demonstrated effectiveness of deep learning models

Improved accuracy and reliability in trash classification

AlphaTrash²⁴

Automated sorting solutions for efficient waste management

Offered practical solutions for automated trash sorting

Improved efficiency and effectiveness in waste management

Augmentation techniques²⁵

Exploration of data augmentation techniques to address data scarcity

Effectiveness of augmentation methods in improving model performance

Enhanced robustness and reliability of trash classification models

Metadata-based approaches²⁶

Utilization of metadata for trash classification

Improved classification accuracy based on metadata information

Enhanced understanding and utilization of metadata in trash classification

Robotics-driven classification²⁷

Trash classification using robotics-driven approaches

Application of robotics for trash classification tasks

Enhanced automation and efficiency in trash classification

CNN models^28,29

Utilization of CNNs for trash classification

Ongoing efforts to improve trash classification methodologies

Continued advancement in trash classification techniques