Portal de Periódicos da CAPES

Analyzing and Design Supporting System for Knit Fabrics. Part 3. Yarn Quantity Ratio Prediction of Jacquard Knit Fabrics Using Image Analysis.

2000; Volume: 53; Issue: 9 Linguagem: Inglês

10.4188/transjtmsj.53.9_t187

1880-1994

Mikiya Kondo, Masayuki Takatera, Yoshio Shimizu,

Industrial Vision Systems and Defect Detection

An analyzing system was developed to predicting the yarn quantity ratio of jacquard knit fabrics automatically without destruction, and the applicability was discussed. The predicting of the yarn quantity of jacquard knit fabrics is very important for the design of knit fabric. To predicting of the yarn quantity ratio of jacquard knit fabrics, computer image processing and regression analysis were implemented in the system. The surface image of knit fabrics was read to a computer with an image scanner. The threshold processing and histogram processing were used to determine the locations of knit stitches. After the detection of stitch loop holes, the regression analysis was performed. In case of the Intersia, Single jacquard, Tube jacquard and Rib jacquard of full needle back stitch, the predicting was performed well. This system can be used instead of the current manual analysis using a balance with destruction samples.