篇名 | Adaptive Gabor Filtering for Fabric Defect Inspection |
---|---|
卷期 | 31:2 |
作者 | Ling Chen 、 Shan Zeng 、 Quansheng Gao 、 Biao Liu |
頁次 | 045-055 |
關鍵字 | detection 、 gabor filter 、 mini-jacquard fabrics 、 PSO 、 EI 、 MEDLINE 、 Scopus |
出刊日期 | 202004 |
DOI | 10.3966/199115992020043102006 |
Based on the characteristics of mini-jacquard fabrics, an adaptive detection and localization method for mini-jacquard fabric by local-local and local-integral texture features is proposed. Gabor filters have been successfully utilized to characterize texture. However, their parameters are often predefined, which may not be optimal for specific tasks such as fabric defect detection. In this paper, we propose to adaptively choose the optimal Gabor filter parameters with the Particle Swarm Optimization (PSO) algorithm. After a defect patch is identified, fuzzy c-means clustering (FCM) algorithm is utilized to locate the defect region. Experimental results demonstrate that the proposed method is able to achieve improved detection accuracy in real-time.