Image classification of cashmere and wool fiber based on LC-KSVD. Issue 1 (June 2021)

• Record Type: Journal Article
• Authors: Sun, Chunhong
• Is Part Of: Journal of physics. Volume 1948:Issue 1(2021)
• Publication Date: 2021-06
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1948/1/012010
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The identification of cashmere and wool fibers has always been a key issue in the textile industry. The classification method of cashmere wool based on microscope image processing technology mainly studies fiber structure characteristics, and the steps are complicated. Affected by changes in breeding methods, breed improvements and climate change, the output of cashmere has decreased, and cashmere fibers have more and more variations, which makes the classification of cashmere wool more difficult. In order to accurately identify cashmere and wool fibers, this paper proposes a multi-feature fusion cashmere wool classification algorithm based on the LC-KSVD algorithm. According to the characteristics of the microscope fiber image, the key information and texture features of the fiber image are extracted, and then the multiple types of texture features are merged into Multi-dimensional feature matrix vector, and finally realize the classification of cashmere wool through the LC-KSVD algorithm. Experiments show that compared with other classification algorithms, the classification accuracy of this method can reach up to 90%, which can be used for subsequent cashmere wool fiber classification and identification.