Investigation on Image-Based Digital Method for Identification on Polyester/Cotton Fiber Category

Lu, Shuaigang; Deng, Na; Xin, Binjie; Wang, Yiliang; Wang, Wenzhen

doi:10.1007/s12221-021-0802-7

Cited by 4 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To achieve the more accurate capability of polyester/ cotton fiber, a novel (Lu et al ( 2021)) [4] properly conceived and obtained suitable digital cross section image analysis framework based on form coefficient analysis was presented. A self-developed microscopic image acquisition system was created to computerize the crossing segment of fiber.…”

Section: Literature Reviewmentioning

confidence: 99%

Development and Application of Information Resources for Education and Teaching of Yarn History Based on 5G Network Technology

Zhang

Yang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Yarn education is a crucial step in producing high-quality textile end products. By offering quick insights into yarn quality, online yarn testing can reduce latency in critical process control, resulting in the manufacture of higher-quality yarns. This paper proposes the development and application of information resources for the education and teaching of yarn history based on fifth-generation (5G) network technologies. Firstly, the yarn database is preprocessed using greyscale transformation and image filtering methods. Secondly, yarn segmentation is implemented using a Multiresolution Markov Random Field (MRMRF) model. Fourier transform and two-scale attention models are used to extract statistical and relative characteristics, respectively. For classifying the yarn photos, we propose using Improved Support Vector Machine (ISVM) classification. The 5G network has been set up, and data is transferred using the Transmission Control Protocol. To enhance the performance, we propose an Improved Particle Swarm Optimization (IPSO) algorithm. The performance of the suggested methodology is analyzed and compared with existing methodologies using the MATLAB simulation tool. The suggested algorithm's classification performance suggests that the 5G framework could provide an accurate, rapid, reliable, and cost-effective solution to industrial automation.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Development and Application of Information Resources for Education and Teaching of Yarn History Based on 5G Network Technology

Zhang

Yang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…There are two main paradigms for fibre detection: traditional image processing-based methods and deep learning-based methods. Traditional image processing-based methods [17] segmented fibres by its spatial statistical features and could be classified into four types, threshold algorithms [4,18], morphological algorithms [8], region-based algorithms [19,20], and edge detection-based algorithms [21,22]. Most of these methods are proposed to deal with fibre overlapping, adhesion and breakage which are the main challenges for fibre detection.…”

Section: Fibre Detectionmentioning

confidence: 99%

“…For instance, Wang et al [5] used the shape features of the triple concentric contours to locate, split, merge and refine fibre contours. Lu et al [21] proposed a composite method which used the edge detection algorithm to identify the external contour, watershed to separate the overlap fibres and an elimination algorithm to remove false fibres. Though efficient, these methods suffer from the inherent difficulty in dealing with the shape diversity of profiled fibres.…”

Section: Fibre Detectionmentioning

confidence: 99%

A weakly supervised learning pipeline for profiled fibre inspection

Chen,

Xiu,

Zheng

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

Automatic profiled fibre recognition and analysis can accelerate quality inspection and contributes to the upgrade of the textile industry. However, these tasks often require significant manual effort to generate instance‐level annotations for fully supervised training. In this paper, the authors propose a weakly supervised pipeline for profiled fibre inspection using electron‐microscopic (EM) images with only image‐level annotations. It automatically identifies fibre instances and estimates shape factors to facilitate fibre quality inspection. As the core of the pipeline, the weakly supervised network (WesNet) is designed to localize hundreds of crowded fibre samples by raw patch generation and fibre sample sifting. Particularly, the composite similarity measurement integrates different patch‐wise similarities, enabling the network to distinguish fibre from background robustly. For quality inspection, the pipeline further analyzes the fibre instances, utilizing several efficient techniques to estimate the shape factors. Experiments on the real fibre electron‐microscopic images demonstrate the efficacy and efficiency of the pipeline. Results show that WesNet outperforms several supervised and weakly supervised methods, including two state‐of‐the‐art weakly supervised networks.

show abstract

“…Binjie Xin’s research team used the traditional method of resin embedding to fix the ramie fiber bundles and proposed an edge-enhanced image processing technology to improve the identification accuracy of fiber edges [ 22 ]. Later, they gained the cross-sectional images of polyester/cotton blended fibers by placing fibers into a suction tube, injecting some prepared resin to embed fibers, and slicing with a microtome [ 23 ]. In 2022, they used a convolutional neural network to identify longitudinally overlapped wool/cashmere fibers [ 24 ].…”

Section: Introductionmentioning

confidence: 99%

A Rapid Quantitative Analysis of Bicomponent Fibers Based on Cross-Sectional In-Situ Observation

Qin

et al. 2023

Polymers

View full text Add to dashboard Cite

To accelerate the industrialization of bicomponent fibers, fiber-based flexible devices, and other technical fibers and to protect the property rights of inventors, it is necessary to develop fast, economical, and easy-to-test methods to provide some guidance for formulating relevant testing standards. A quantitative method based on cross-sectional in-situ observation and image processing was developed in this study. First, the cross-sections of the fibers were rapidly prepared by the non-embedding method. Then, transmission and reflection metallographic microscopes were used for in-situ observation and to capture the cross-section images of fibers. This in-situ observation allows for the rapid identification of the type and spatial distribution structure of the bicomponent fiber. Finally, the mass percentage content of each component was calculated rapidly by AI software according to its density, cross-section area, and total test samples of each component. By comparing the ultra-depth of field microscope, differential scanning calorimetry (DSC), and chemical dissolution method, the quantitative analysis was fast, accurate, economical, simple to operate, energy-saving, and environmentally friendly. This method will be widely used in the intelligent qualitative identification and quantitative analysis of bicomponent fibers, fiber-based flexible devices, and blended textiles.

show abstract

Investigation on Image-Based Digital Method for Identification on Polyester/Cotton Fiber Category

Cited by 4 publications

References 26 publications

Development and Application of Information Resources for Education and Teaching of Yarn History Based on 5G Network Technology

Development and Application of Information Resources for Education and Teaching of Yarn History Based on 5G Network Technology

A weakly supervised learning pipeline for profiled fibre inspection

A Rapid Quantitative Analysis of Bicomponent Fibers Based on Cross-Sectional In-Situ Observation

Contact Info

Product

Resources

About