Title :
Spectral and spatial domain approach for fabric defect detection and classification
Author :
Sakhare, Kaustubh ; Kulkarni, Akshada ; Kumbhakarn, Mansi ; Kare, Nachiket
Author_Institution :
Electron. & Telecommun., Pune Inst. of Comput. Technol., Pune, India
Abstract :
Fabric inspection has an important role of preventing delivery of inferior quality product. The fabric defect inspection process is traditionally carried out with human visual inspection that proves to be insufficient and costly. Hence in order to reduce the cost and wastage of time, Computer vision based fabric defect detection is required. Robust and efficient fabric defect detection algorithms are required for inspection. This paper presents spectral domain and spatial domain approach for fabric defect detection. Spectral domain approach includes Fast Fourier Transform (FFT), Discrete Cosine Transform (DCT), wavelet and Gabor filter. Spatial domain approach includes two methods morphological operations and statistical operations.
Keywords :
Gabor filters; automatic optical inspection; computer vision; discrete cosine transforms; fabrics; fast Fourier transforms; image classification; image filtering; product quality; statistical analysis; wavelet transforms; DCT; FFT; Gabor filter; computer vision based fabric defect detection; discrete cosine transform; fabric defect detection algorithms; fast Fourier transform; inferior quality product; morphological operations; spatial domain approach; spectral domain approach; statistical operations; wavelet filter; Cameras; Discrete cosine transforms; Fabrics; Lead; Propulsion; Reliability; DFT; FFT; Fabric defects; Gabor filtering; Wavelet; defect detection and classification; morpholog-ical operations; statistical analysis;
Conference_Titel :
Industrial Instrumentation and Control (ICIC), 2015 International Conference on
Conference_Location :
Pune
DOI :
10.1109/IIC.2015.7150820
