Automatic System to Control the Anisotropy of Lyocell Felted Webs


The webs produced using a card as former system generally keep the fiber orientation. This effect is not very important for a great number of applications. But if an even pore distribution is necessary or the mechanical properties are to be uniform in all directions, then this system doesn’t work very well. Taking into account these considerations, a new technological solution has been specifically devised to control the fibre orientation during the drafting operation in pre-needled felts. A prototype is used as drafting unit with the rolls driven by stepper motors and controlled by computer. Two video cameras collect images of the surface of the web,  before and after the drafting operation,   and process them in a computer. The values of the images textural descriptors, are calculated and compared. It is intended an increase in entropy which means that the web become less ordered and this is the situation we are aiming for. A computer program will adjust automatically the speed and pressure of the drafting cylinders so as to achieve the best possible situation in terms of the isotropy of the final product and, consequently, MD:CD ratio close to 1.

Keywords: Anisotropy, Nonwoven felts, Image analysis, Textural descriptors, Estiragem

[1] Mário F.N., “Novo Método de Controlo na Produção de Têxteis Técnicos”, XX Seminário da APETT, Famalicão, 1999;

[2] Belino N.J.R., “Optimização de um sistema automático de controlo da isotropia de mantos pré- agulhados por análise de textura e recuperação de imagem baseada em conteúdo, PhD Thesis, UBI, 2006

[3] Tsai, P.P., “Theoretical and Experimental Investigation on the Relationship Between the Nonwoven Structure and the Web Properties”, International Nonwovens Journal, N°4. pp. 33- 36,2002

[4] Hearle, J.W.S., Stevenson, P.J. “A Study of needled Fabrics, Part III: “The Anisotropy of Nonwoven Fabrics”, Textile Research Journal, n°11, pp. 877-888, 1963

[5] Schalkoff, R., “Digital Image Processing and Computer Vision”, John Wiley & Sons, Inc., 1989

[6] Baraldi, A., Parmigianni, F., “An Investigation of Textural Characteristics Associated With Grey-Level Co-occurrence Matrix Statistical Parameters”, IEEE, Vol. 33, N°. 2, pp. 195-199, 1995

[7] Haralick, R M., Shanmungam, K., Dinstein,I., “Textural Features for Image Classification”, IEEE, Vol. 3, N°. 6, pp. 610-621, 1973.