Related Experiment Video
Updated: May 14, 2026
06:02
Disentangling High Strength Copolymer Aramid Fibers to Enable the Determination of Their Mechanical Properties
Published on: September 1, 2018
7.2K
Hybrid feature-based machine vision method for objective evaluation of textile pilling and fuzzing
Qingchun Jiao1, Zifan Qian1, Yue Dong2
1School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Hangzhou, China.
Plos One
|September 3, 2025
Summary
This study introduces a machine vision method using a Hybrid Feature-based Depthwise Separable Attention Network (HDAN-PF) for objective textile pilling and fuzzing evaluation. The novel approach achieves 96.26% accuracy, improving textile quality assessment.
Area of Science:
- Textile Science
- Computer Vision
- Machine Learning
Background:
- Textile pilling and fuzzing are critical quality indicators.
- Current subjective evaluation methods are inefficient and prone to errors.
- Objective assessment is needed for consistent textile quality control.
Purpose of the Study:
- To develop an objective, efficient, and accurate method for evaluating textile pilling and fuzzing grades.
- To introduce a novel machine vision approach for textile quality assessment.
- To overcome the limitations of subjective grading systems.
Main Methods:
- Proposed a Hybrid Feature-based Depthwise Separable Attention Network (HDAN-PF).
- Employed a machine vision approach integrating frequency and space domain features.
- Utilized Channel Attention and Depthwise Separable Convolutions for feature extraction and efficiency.
Main Results:
- Achieved a classification accuracy of 96.26% on diverse fabric images.
- Demonstrated robust generalization capabilities across various fabric types.
- The HDAN-PF model effectively fuses multi-domain features for precise evaluation.
Conclusions:
- The proposed Hybrid Feature-based Machine Vision Method offers a transformative solution for objective textile pilling and fuzzing assessment.
- This approach enhances consistency and efficiency in textile quality evaluation.
- The method shows significant practical utility for the textile industry.
