Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables

Yeong Hyeon Park; Myung Jin Kim; Won Seok Park; Juneho Yi

doi:10.1109/SIST58284.2023.10223525

Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables

Yeong Hyeon Park
, Myung Jin Kim
, Won Seok Park
, Juneho Yi

Department of Electronic and Electrical Engineering

Sungkyunkwan University
SK Corporation

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Automated classification of recyclable waste is necessary to process a huge amount of recyclables for reuse. This research features recycling a pre-trained attention mechanism for cropping region of interest (RoI) for efficient classification of recyclable waste. We report that an attention mechanism pre-trained with the MNIST dataset, followed by simple morphological operations, successfully provides a bounding box for a recyclable object to be fed into object recognition models such as ResNet50 and EffNetB0. This way, we avoid the cost of annotating large datasets to train state-of-the-art object detection models such as YOLO and R-CNN. Experimental results using the Recyclable Solid Waste Dataset (RSWD) show that our attention-based RoI cropping method is effective enough to separate an object for recognition to achieve accurate classification of recyclables.

Original language	English
Title of host publication	SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	429-433
Number of pages	5
ISBN (Electronic)	9798350335040
DOIs	https://doi.org/10.1109/SIST58284.2023.10223525
State	Published - 2023
Event	2023 IEEE International Conference on Smart Information Systems and Technologies, SIST 2023 - Astana, Kazakhstan Duration: 4 May 2023 → 6 May 2023

Publication series

Name	SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings

Conference

Conference	2023 IEEE International Conference on Smart Information Systems and Technologies, SIST 2023
Country/Territory	Kazakhstan
City	Astana
Period	4/05/23 → 6/05/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities
SDG 12 Responsible Consumption and Production

Keywords

attention mechanism
recycling
RoI detection
system automation
waste classification

Access to Document

10.1109/SIST58284.2023.10223525

Cite this

Park, Y. H., Kim, M. J., Park, W. S., & Yi, J. (2023). Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables. In SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings (pp. 429-433). (SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SIST58284.2023.10223525

Park, Yeong Hyeon ; Kim, Myung Jin ; Park, Won Seok et al. / Recycling for Recycling : RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables. SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 429-433 (SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings).

@inproceedings{3c88432dc52a41f9b7167df8bce52126,

title = "Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables",

abstract = "Automated classification of recyclable waste is necessary to process a huge amount of recyclables for reuse. This research features recycling a pre-trained attention mechanism for cropping region of interest (RoI) for efficient classification of recyclable waste. We report that an attention mechanism pre-trained with the MNIST dataset, followed by simple morphological operations, successfully provides a bounding box for a recyclable object to be fed into object recognition models such as ResNet50 and EffNetB0. This way, we avoid the cost of annotating large datasets to train state-of-the-art object detection models such as YOLO and R-CNN. Experimental results using the Recyclable Solid Waste Dataset (RSWD) show that our attention-based RoI cropping method is effective enough to separate an object for recognition to achieve accurate classification of recyclables.",

keywords = "attention mechanism, recycling, RoI detection, system automation, waste classification",

author = "Park, \{Yeong Hyeon\} and Kim, \{Myung Jin\} and Park, \{Won Seok\} and Juneho Yi",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Conference on Smart Information Systems and Technologies, SIST 2023 ; Conference date: 04-05-2023 Through 06-05-2023",

year = "2023",

doi = "10.1109/SIST58284.2023.10223525",

language = "English",

series = "SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "429--433",

booktitle = "SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings",

}

Park, YH, Kim, MJ, Park, WS & Yi, J 2023, Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables. in SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings. SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 429-433, 2023 IEEE International Conference on Smart Information Systems and Technologies, SIST 2023, Astana, Kazakhstan, 4/05/23. https://doi.org/10.1109/SIST58284.2023.10223525

Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables. / Park, Yeong Hyeon; Kim, Myung Jin; Park, Won Seok et al.
SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2023. p. 429-433 (SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Recycling for Recycling

T2 - 2023 IEEE International Conference on Smart Information Systems and Technologies, SIST 2023

AU - Park, Yeong Hyeon

AU - Kim, Myung Jin

AU - Park, Won Seok

AU - Yi, Juneho

PY - 2023

Y1 - 2023

N2 - Automated classification of recyclable waste is necessary to process a huge amount of recyclables for reuse. This research features recycling a pre-trained attention mechanism for cropping region of interest (RoI) for efficient classification of recyclable waste. We report that an attention mechanism pre-trained with the MNIST dataset, followed by simple morphological operations, successfully provides a bounding box for a recyclable object to be fed into object recognition models such as ResNet50 and EffNetB0. This way, we avoid the cost of annotating large datasets to train state-of-the-art object detection models such as YOLO and R-CNN. Experimental results using the Recyclable Solid Waste Dataset (RSWD) show that our attention-based RoI cropping method is effective enough to separate an object for recognition to achieve accurate classification of recyclables.

AB - Automated classification of recyclable waste is necessary to process a huge amount of recyclables for reuse. This research features recycling a pre-trained attention mechanism for cropping region of interest (RoI) for efficient classification of recyclable waste. We report that an attention mechanism pre-trained with the MNIST dataset, followed by simple morphological operations, successfully provides a bounding box for a recyclable object to be fed into object recognition models such as ResNet50 and EffNetB0. This way, we avoid the cost of annotating large datasets to train state-of-the-art object detection models such as YOLO and R-CNN. Experimental results using the Recyclable Solid Waste Dataset (RSWD) show that our attention-based RoI cropping method is effective enough to separate an object for recognition to achieve accurate classification of recyclables.

KW - attention mechanism

KW - recycling

KW - RoI detection

KW - system automation

KW - waste classification

UR - https://www.scopus.com/pages/publications/85171970331

U2 - 10.1109/SIST58284.2023.10223525

DO - 10.1109/SIST58284.2023.10223525

M3 - Conference contribution

AN - SCOPUS:85171970331

T3 - SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings

SP - 429

EP - 433

BT - SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 4 May 2023 through 6 May 2023

ER -

Park YH, Kim MJ, Park WS, Yi J. Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables. In SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2023. p. 429-433. (SIST 2023 - 2023 IEEE International Conference on Smart Information Systems and Technologies, Proceedings). doi: 10.1109/SIST58284.2023.10223525

Recycling for Recycling: RoI Cropping by Recycling a Pre-Trained Attention Mechanism for Accurate Classification of Recyclables

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this