Authors: Yu-Tang Lee; Chung Yeh
Addresses: Department of Industrial Engineering and Management, Chaoyang University of Technology, 168, Jifeng E. Rd., Wufeng District, Taichung, 41349, Taiwan ' Department of Industrial Engineering and Systems Management, Feng Chia University, Taichung City, 407, Taiwan
Abstract: Various defects existed on the surface of calf leather could affect its usable area and the salable price. No international criterion specifies the compensatory credits for calf leather surface defects which cause additional cost between supplier and purchaser in complicated negotiation process. This paper is to develop an artificial intellectual technique to implement the automatic recognition for types of leather defect and to compensate for leather defective unusable area in order to bridge trading gap. Data of calf defects from sample is extracted to develop an automatic recognition system via artificial intellectual techniques – ANN learning process is introduced to make a sustainable automatic recognition system used to identify types of categories for upcoming leathers under inspection, business transaction; the mean error rate of recognising leather defect is less than 2.16% and the mean deviation rate for compensation area is 0.03% under this simulated transaction.
Keywords: leather surface defects; artificial neural network; ANN; digit image processing; mean error rate of recognising leather defect; mean deviation rate for the leather area.
International Journal of Information Technology and Management, 2020 Vol.19 No.2/3, pp.93 - 117
Received: 12 Jul 2017
Accepted: 24 Oct 2017
Published online: 02 Apr 2020 *