Multi-region automatic identification of jacquard warp knitted fabric weave based on wavelet transformation

The aim of this study is to simplify the regional design process of jacquard regions of warp knitted fabrics.In view of the texture features in a number of regions,in particular,the difficulty of mesh region classification,an automatic regional identification method is proposed.Firstly median filtering is used for noise elimination in the original image.Secondly,wavelet transform method is adopted to extract different regional texture features of the fabric.finally,when the number of category is known,K means clustering method is used to segment these texture features,and thus automatic regional identification of fabric weave is realized.Experiments show that the method has a good identification effect.Jacquard warp knitted fabric with two or more regions can be distinguished in supervised and automatic manner,and it is adapted to the automatic segmentation of the mesh regions,with better regional consistency and edge accuracy.