EcoReward: An Automated Deep Learning Framework for Recyclable Plastic Classification and Incentive Distribution Using YOLOv8

The growing consumption of plastic-based products and packaging poses significant challenges for municipal solid waste management. Conventional recycling workflows often depend on manual identification and segregation, which can cause delays, inconsistent classification, and contamination of recyclable streams, while the absence of immediate and transparent incentives may limit public participation. This paper presents EcoReward, an automated deep learning framework that couples recyclable-packaging detection with an incentive distribution mechanism. The system uses a custom-trained YOLOv8n detector to identify four packaging categories: PET bottles, HDPE containers, squeeze tubes, and UHT beverage cartons/multilayer packaging. A web-based interface and a Python FastAPI backend support image submission, object detection, visual annotation, reward calculation, and transaction recording. To suit limited cloud resources, the implementation applies client-side image resizing with HTML5 Canvas, CPU-based inference, and explicit memory management. Detected objects are passed to a class-dependent reward allocation model that awards Eco-Points according to the number and category of recognized items, and each transaction (detected counts, reward value, and timestamp) is stored in SQLite. Unlike image-level classification, EcoReward performs object-level detection and counting and links the output to a structured incentive workflow. The paper describes the architecture, theoretical background, mathematical reward model, implementation observations, and ethical considerations. Quantitative evaluation of detection performance on a documented test set is reserved for future work.

Authors

Institutions

Publication Details

Journal
International Journal of Scientific Research in Science Engineering and Technology
Published
2026-11-18
DOI
https://doi.org/10.32628/ijsrset2613610
Primary Topic
Advanced Neural Network Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

EcoReward: An Automated Deep Learning Framework for Recyclable Plastic Classification and Incentive Distribution Using YOLOv8

Arati Dandavate, Tanishq Kudale, Vedant Shete, Samarth Lokhande
4 links
International Journal of Scientific Research in Science Engineering and Technology
Advanced Neural Network Applications
article

EcoReward: An Automated Deep Learning Framework for Recyclable Plastic Classification and Incentive Distribution Using YOLOv8

Arati Dandavate, Tanishq Kudale, Vedant Shete, Samarth Lokhande
article en

Abstract

The growing consumption of plastic-based products and packaging poses significant challenges for municipal solid waste management. Conventional recycling workflows often depend on manual identification and segregation, which can cause delays, inconsistent classification, and contamination of recyclable streams, while the absence of immediate and transparent incentives may limit public participation. This paper presents EcoReward, an automated deep learning framework that couples recyclable-packaging detection with an incentive distribution mechanism. The system uses a custom-trained YOLOv8n detector to identify four packaging categories: PET bottles, HDPE containers, squeeze tubes, and UHT beverage cartons/multilayer packaging. A web-based interface and a Python FastAPI backend support image submission, object detection, visual annotation, reward calculation, and transaction recording. To suit limited cloud resources, the implementation applies client-side image resizing with HTML5 Canvas, CPU-based inference, and explicit memory management. Detected objects are passed to a class-dependent reward allocation model that awards Eco-Points according to the number and category of recognized items, and each transaction (detected counts, reward value, and timestamp) is stored in SQLite. Unlike image-level classification, EcoReward performs object-level detection and counting and links the output to a structured incentive workflow. The paper describes the architecture, theoretical background, mathematical reward model, implementation observations, and ethical considerations. Quantitative evaluation of detection performance on a documented test set is reserved for future work.

International Journal of Scientific Research in Science Engineering and TechnologyVol. 13(6)
MIT Art, Design and Technology University (IN)
Responsible consumption and production, Industry, innovation and infrastructure
Advanced Neural Network Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon