We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Data Augmentation Using Background Replacement for Automated Sorting of Littered Waste
Summary
This paper describes using data augmentation techniques to improve machine learning models for automated sorting of litter in outdoor environments. Better waste sorting technology could improve plastic recycling rates and reduce the amount of plastic that ends up fragmenting into microplastics.
The introduction of sophisticated waste treatment plants is making the process of trash sorting and recycling more and more effective and eco-friendly. Studies on Automated Waste Sorting (AWS) are greatly contributing to making the whole recycling process more efficient. However, a relevant issue, which remains unsolved, is how to deal with the large amount of waste that is littered in the environment instead of being collected properly. In this paper, we introduce BackRep: a method for building waste recognizers that can be used for identifying and sorting littered waste directly where it is found. BackRep consists of a data-augmentation procedure, which expands existing datasets by cropping solid waste in images taken on a uniform (white) background and superimposing it on more realistic backgrounds. For our purpose, realistic backgrounds are those representing places where solid waste is usually littered. To experiment with our data-augmentation procedure, we produced a new dataset in realistic settings. We observed that waste recognizers trained on augmented data actually outperform those trained on existing datasets. Hence, our data-augmentation procedure seems a viable approach to support the development of waste recognizers for urban and wild environments.
Sign in to start a discussion.
More Papers Like This
Microplastics in the rough: using data augmentation to identify plastics contaminated by water and plant matter
This study developed machine learning approaches using data augmentation to improve the identification of microplastics in "real world" samples where particles are contaminated by water droplets, soil, or plant material. Accurately classifying weathered and dirty microplastics from spectral images is a practical challenge that limits field research, and the techniques developed here improve detection accuracy. Better identification tools are a necessary step toward reliable monitoring of microplastic pollution across diverse environments.
Enhancing Waste Management with a Deep Learning-based Automatic Garbage Classifier
This paper is not about microplastics; it presents a deep learning convolutional neural network system for automatically classifying garbage by material type to improve waste sorting efficiency and reduce the labor burden of manual waste management.
A Machine Arm to Assist in Trash Sorting using machine Learning and Object Detection
Not relevant to microplastics — this paper describes a robotic arm system that uses machine learning and computer vision to sort recyclable waste materials, focused on automation of waste sorting processes.
A Smart Garbage Classification based on Deep Learning
Researchers developed an AI-powered garbage classification system using deep learning to automatically sort waste categories. Accurate automated waste sorting could improve plastic recycling rates, reducing the amount of plastic that eventually breaks down into environmental microplastics.
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.