Next Article in Journal
Simulation-Based Hybrid Energy Storage Composite-Target Planning with Power Quality Improvements for Integrated Energy Systems in Large-Building Microgrids
Previous Article in Journal
A Resource Allocation Algorithm for Cloud-Network Collaborative Satellite Networks with Differentiated QoS Requirements
Previous Article in Special Issue
Multi-Modular Network-Based Retinex Fusion Approach for Low-Light Image Enhancement
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

Efficient Fabric Classification and Object Detection Using YOLOv10

1
Department of Software Convergence, Soonchunhyang University, Asan-si 31538, Republic of Korea
2
Department of Computer Science, Kennesaw State University, Marietta, GA 30060, USA
3
Department of Computer Software Engineering, Soonchunhyang University, Asan-si 31538, Republic of Korea
*
Author to whom correspondence should be addressed.
Electronics 2024, 13(19), 3840; https://doi.org/10.3390/electronics13193840 (registering DOI)
Submission received: 29 August 2024 / Revised: 24 September 2024 / Accepted: 27 September 2024 / Published: 28 September 2024
(This article belongs to the Special Issue Modern Computer Vision and Image Analysis)

Abstract

The YOLO (You Only Look Once) series is renowned for its real-time object detection capabilities in images and videos. It is highly relevant in industries like textiles, where speed and accuracy are critical. In the textile industry, accurate fabric type detection and classification are essential for improving quality control, optimizing inventory management, and enhancing customer satisfaction. This paper proposes a new approach using the YOLOv10 model, which offers enhanced detection accuracy, processing speed, and detection on the torn path of each type of fabric. We developed and utilized a specialized, annotated dataset featuring diverse textile samples, including cotton, hanbok, cotton yarn-dyed, and cotton blend plain fabrics, to detect the torn path in fabric. The YOLOv10 model was selected for its superior performance, leveraging advancements in deep learning architecture and applying data augmentation techniques to improve adaptability and generalization to the various textile patterns and textures. Through comprehensive experiments, we demonstrate the effectiveness of YOLOv10, which achieved an accuracy of 85.6% and outperformed previous YOLO variants in both precision and processing speed. Specifically, YOLOv10 showed a 2.4% improvement over YOLOv9, 1.8% over YOLOv8, 6.8% over YOLOv7, 5.6% over YOLOv6, and 6.2% over YOLOv5. These results underscore the significant potential of YOLOv10 in automating fabric detection processes, thereby enhancing operational efficiency and productivity in textile manufacturing and retail.
Keywords: object detection; computer vision; convolutional neural networks; deep learning; fabric classification; YOLO object detection; computer vision; convolutional neural networks; deep learning; fabric classification; YOLO

Share and Cite

MDPI and ACS Style

Mao, M.; Lee, A.; Hong, M. Efficient Fabric Classification and Object Detection Using YOLOv10. Electronics 2024, 13, 3840. https://doi.org/10.3390/electronics13193840

AMA Style

Mao M, Lee A, Hong M. Efficient Fabric Classification and Object Detection Using YOLOv10. Electronics. 2024; 13(19):3840. https://doi.org/10.3390/electronics13193840

Chicago/Turabian Style

Mao, Makara, Ahyoung Lee, and Min Hong. 2024. "Efficient Fabric Classification and Object Detection Using YOLOv10" Electronics 13, no. 19: 3840. https://doi.org/10.3390/electronics13193840

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop