Paper
19 October 2023 Textile fabric image classification method based on enhanced deep learning features
ShiCong Cai, WanRu Zhu, KeTong Chen, Huabiao Qin
Author Affiliations +
Proceedings Volume 12709, Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023); 127094L (2023) https://doi.org/10.1117/12.2684999
Event: Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023), 2023, Nanjing, China
Abstract
In recent years, the deep learning features, which was extracted and coded by neural network, have been widely used in image classification and achieved significant results. Textile fabric industry usually classify the items according to image patterns of fabrics, in which the shape and color of image have more distinguish effect. So how to use the image characteristics of textile fabrics to enhance the discrimination of deep learning features is the key process to improve the accuracy of classification of textile fabric. This paper proposes a method of textile fabric classification based on the deep learning features enhanced by Histograms of Oriented Gradients (HOG) features and HS histogram statistical features extracted in HSV domain. By fusing more image shape and color information with deep learning features, the enhanced deep learning features have greater discrimination. The experiment shows that the proposed method has good classification result, achieve 92.4% accuracy on the corresponding dataset.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
ShiCong Cai, WanRu Zhu, KeTong Chen, and Huabiao Qin "Textile fabric image classification method based on enhanced deep learning features", Proc. SPIE 12709, Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023), 127094L (19 October 2023); https://doi.org/10.1117/12.2684999
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Deep learning

Histograms of oriented gradient

Feature extraction

Image enhancement

Image classification

RGB color model

Color

Back to Top