http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
이정범(Jungbeom Lee),정다흰(Dahuin Jung),윤성로(Sungroh Yoon) 대한전자공학회 2018 대한전자공학회 학술대회 Vol.2018 No.11
Measuring the quality of clothes and fabrics is an indispensable factor in the respect to the satisfactory of consumers and the profitability for the apparel companies. We propose a method to automatically detect the flaws in fabric, which is the main material of apparel products, without human involvement. In the absence of data on defect samples, synthetic abnormal data is generated using classical method. Using the synthesized data, we train convolutional neural network (CNN) to determine a given patch contains defects or not. The fabric is inspected whether or not it contains defects through trained CNN. In this paper, it is shown that deep learning can be applied to practical applications related to the clothing industry.