Cellulose Materials Identification: The Effect of Dimensionality of Colour Photography Data

This paper describes a simple rapid staining microcolorimetric method for analytical fibre material identification using colour vectors of stained fibre material photography. The number of morphological characteristics (nM), number of stains (nS), colour information dimensionality (nDC), and picture elementary points number (npx) can play a key role in distinguishing fibre materials, correct identification, discriminatory power dP (%), and efficacy. Experiments were performed to achieve the most accurate results with a minimum volume of data; the dimensionality reduction was made experimentally by setting nM = 0, nS = 1, nDC , and the effect of number of pixels on the dP (%) was measured. The correct identification was achieved by less than 100 pixels when using 2 colour vectors, and by less than 50 pixels when using 3 colour vectors: R, G, and B. The real area of the pixels used for correct identification was less than 0.1 mm2 in the used model system of the cellulose fibre materials.

[1]  Jozef Sima,et al.  Cellulose Fibre Identification through Color Vectors of Stained Fibre , 2015 .

[2]  Ernestina Menasalvas Ruiz,et al.  A methodology to compare Dimensionality Reduction algorithms in terms of loss of quality , 2014, Inf. Sci..

[3]  Peyman Adibi,et al.  Two-stage multiple kernel learning for supervised dimensionality reduction , 2015, Pattern Recognit..

[4]  Metin Nafi Gürcan,et al.  Microscopic image analysis for quantitative characterization of muscle fiber type composition , 2011, Comput. Medical Imaging Graph..

[5]  P. Whittaker,et al.  Fibrin architecture in clots: a quantitative polarized light microscopy analysis. , 2009, Blood cells, molecules & diseases.

[6]  Kyoungok Kim,et al.  Sentiment visualization and classification via semi-supervised nonlinear dimensionality reduction , 2014, Pattern Recognit..

[7]  N. Gunde-Cimerman,et al.  FTIR spectroscopy of biodegraded historical textiles , 2011 .

[8]  Lin Yan,et al.  Improved time series land cover classification by missing-observation-adaptive nonlinear dimensionality reduction , 2015 .

[9]  I. Rezić,et al.  Simple methods for characterization of metals in historical textile threads. , 2010, Talanta.

[10]  Tao Li,et al.  Technical investigation of 15th and 19th century Chinese paper currencies: Fiber use and pigment identification , 2013 .

[11]  David Legland,et al.  Automated clustering of lignocellulosic fibres based on morphometric features and using clustering of variables , 2013 .

[12]  R. E. Raj,et al.  Characterization of new natural cellulosic fiber from Cissus quadrangularis stem. , 2015, Carbohydrate polymers.

[13]  N. R. Sakthivel,et al.  Comparison of dimensionality reduction techniques for the fault diagnosis of mono block centrifugal pump using vibration signals , 2014 .

[14]  Zhulan Liu,et al.  The Utilization of Soybean Straw. I. Fiber Morphology and Chemical Characteristics , 2015 .

[15]  Manfred Burghammer,et al.  Identification of ancient textile fibres from Khirbet Qumran caves using synchrotron radiation microbeam diffraction , 2004 .

[16]  A. Lattuati-Derieux,et al.  Analysis of a varnish applied by Nicéphore Niépce to make a print used for photographic purpose transparent , 2010 .

[17]  Mohammad Jawaid,et al.  Cell wall ultrastructure, anatomy, lignin distribution, and chemical composition of Malaysian cultivated kenaf fiber , 2010 .

[18]  Feng Zhao,et al.  Identification of ancient textiles from Yingpan, Xinjiang, by multiple analytical techniques , 2011 .