Multichannel Convolutional Networks for Classification and Localization of Construction and Demolition Waste

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21110%2F25%3A00384947" target="_blank" >RIV/68407700:21110/25:00384947 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-94600-4_3" target="_blank" >https://doi.org/10.1007/978-3-031-94600-4_3</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-94600-4_3" target="_blank" >10.1007/978-3-031-94600-4_3</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Multichannel Convolutional Networks for Classification and Localization of Construction and Demolition Waste

Popis výsledku v původním jazyce

Effective management of construction and demolition waste (CDW) remains a critical concern for environmental sustainability and economic efficiency. Inadequate sorting mechanisms result in subopti-mal resource use and diminished recycling opportunities. This paper presents a sophisticated machine-learning approach that utilizes multi-channel convolutional neural networks (CNNs) to improve CDW sorting and recycling. Our methodology incorporates a two-phase deep learning model, beginning with a U-Net architecture for detailed, pixel-level seg-mentation of waste materials. This phase achieves an Intersection over Union (IoU) of 0.902, ensuring precise material localization. Following this, a CNN based on EfficientNet is applied, achieving a classification accuracy of 99%, demonstrating significant improvement over traditional methods. Our comparative analysis highlights the practical advantages of this approach, including labor savings and superior classification accu-racy compared to manual sorting methods. By focusing on integrat-ing real-time processing capabilities, this research provides actionable insights for advancing automated CDW management systems. Addition-ally, providing codes, training and testing datasets, and supplementary resources facilitates further research and implementation. These find-ings underline the potential of machine learning to revolutionize CDW sorting, aligning with the principles of environmental sustainability and economic efficiency.

Název v anglickém jazyce

Multichannel Convolutional Networks for Classification and Localization of Construction and Demolition Waste

Popis výsledku anglicky

Effective management of construction and demolition waste (CDW) remains a critical concern for environmental sustainability and economic efficiency. Inadequate sorting mechanisms result in subopti-mal resource use and diminished recycling opportunities. This paper presents a sophisticated machine-learning approach that utilizes multi-channel convolutional neural networks (CNNs) to improve CDW sorting and recycling. Our methodology incorporates a two-phase deep learning model, beginning with a U-Net architecture for detailed, pixel-level seg-mentation of waste materials. This phase achieves an Intersection over Union (IoU) of 0.902, ensuring precise material localization. Following this, a CNN based on EfficientNet is applied, achieving a classification accuracy of 99%, demonstrating significant improvement over traditional methods. Our comparative analysis highlights the practical advantages of this approach, including labor savings and superior classification accu-racy compared to manual sorting methods. By focusing on integrat-ing real-time processing capabilities, this research provides actionable insights for advancing automated CDW management systems. Addition-ally, providing codes, training and testing datasets, and supplementary resources facilitates further research and implementation. These find-ings underline the potential of machine learning to revolutionize CDW sorting, aligning with the principles of environmental sustainability and economic efficiency.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20101 - Civil engineering

Projekt

<a href="/cs/project/EH22_008%2F0004590" target="_blank" >EH22_008/0004590: Robotika a pokročilá průmyslová výroba</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Futuristic Computational Systems and Advanced Engineering for the Society, Proceedings of the 6th International Conference on Artificial Intelligence and Applied Mathematics in Engineering ICAIAME 2024, Volume 2

ISBN

9783031946004

ISSN

2731-5010

e-ISSN

—

Počet stran výsledku

17

Strana od-do

26-42

Název nakladatele

Springer Nature Switzerland AG

Místo vydání

Basel

Místo konání akce

Warsaw

Datum konání akce

26. 9. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—