We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Efficient algorithmic coupling technique for precision recycling of seven types of mixed plastic waste
Summary
Researchers developed a two-step machine learning coupling technique combining Linear Support Vector Classification (Linear-SVC) with a Multi-layer Perceptron (MLP) to improve the precision of sorting seven types of mixed plastic waste. The coupling technique raised overall plastic identification accuracy from 94.7% to 97.7% and substantially improved classification accuracy for HDPE and LDPE from 79-94%, while also reducing computation time compared to the single-step MLP approach.
Abstract The annual global production of plastic waste, characterized by complex composition and challenges in separation, necessitates immediate and comprehensive measures for the recycling and disposal of mixed plastic waste in an environmentally friendly and meticulous manner. This study introduces an efficient two-step coupling technique, employing Linear Support Vector Classification (Linear-SVC) in tandem with Multi-layer Perceptron (MLP). The application of this coupling technique elevates the overall accuracy of identifying seven types of plastics from 94.7% to an impressive 97.7%. Furthermore, the method exhibits a reduced running time compared to the one-step method of MLP. Notably, the classification accuracy for high-density polyethylene (HDPE) and low-density polyethylene (LDPE) experiences a substantial improvement from 79–94%, outperforming the one-step MLP method. This coupling technique emerges as an effective strategy, contributing significantly to the harmless and precise recycling of waste plastics.
Sign in to start a discussion.
More Papers Like This
Deep transfer learning benchmark for plastic waste classification
Researchers benchmarked six deep transfer learning models for classifying plastic waste types, achieving high accuracy in automated sorting that could help address plastic pollution by improving recycling efficiency.
Advancing Plastic Waste Classification and Recycling Efficiency: Integrating Image Sensors and Deep Learning Algorithms
Researchers developed a deep learning approach combined with image sensors to improve plastic waste classification and recycling efficiency. The study demonstrates that this method can distinguish between chemically similar plastics like PET and PET-G that conventional near-infrared spectroscopy struggles to differentiate, potentially improving automated sorting systems.
Application of AI-Enabled Computer Vision Technology for Segregation of Industrial Plastic Wastes
Researchers developed an AI-powered computer vision system to segregate mixed industrial plastic wastes by polymer type, addressing a key barrier to effective plastic recycling. The system achieved high classification accuracy across common plastic categories, demonstrating that machine vision can improve sorting efficiency and recycled plastic quality.
Mid-infrared spectroscopy and machine learning for postconsumer plastics recycling
Mid-infrared spectroscopy combined with machine learning was developed to sort and identify postconsumer plastics, aiming to prevent contamination and improve recycling stream purity. The approach could help close material loops and reduce the volume of plastic ultimately entering the environment.
An Automatic Garbage Classification System Based on Deep Learning
Researchers developed an automated garbage classification system using a deep learning algorithm based on ResNet-34, achieving 99% classification accuracy with a processing time of under one second per item. Automated waste sorting technology like this could improve the efficiency of plastic waste recovery and reduce mismanaged plastic that eventually becomes environmental pollution.