WO2023222945A1 - Estimating the viscosity of textile materials - Google Patents
Estimating the viscosity of textile materials Download PDFInfo
- Publication number
- WO2023222945A1 WO2023222945A1 PCT/FI2023/050264 FI2023050264W WO2023222945A1 WO 2023222945 A1 WO2023222945 A1 WO 2023222945A1 FI 2023050264 W FI2023050264 W FI 2023050264W WO 2023222945 A1 WO2023222945 A1 WO 2023222945A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- cellulose
- calibration
- based textile
- textile sample
- viscosity
- Prior art date
Links
- 239000004753 textile Substances 0.000 title claims abstract description 134
- 239000000463 material Substances 0.000 title description 16
- 229920002678 cellulose Polymers 0.000 claims abstract description 111
- 239000001913 cellulose Substances 0.000 claims abstract description 111
- 238000013178 mathematical model Methods 0.000 claims abstract description 65
- 238000001228 spectrum Methods 0.000 claims abstract description 65
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 238000004590 computer program Methods 0.000 claims description 28
- 239000000126 substance Substances 0.000 claims description 15
- 238000013145 classification model Methods 0.000 claims description 14
- 239000000203 mixture Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 description 31
- 238000004458 analytical method Methods 0.000 description 13
- 238000012360 testing method Methods 0.000 description 12
- 230000003595 spectral effect Effects 0.000 description 9
- 238000000701 chemical imaging Methods 0.000 description 8
- 238000002835 absorbance Methods 0.000 description 7
- 239000004744 fabric Substances 0.000 description 6
- 238000004064 recycling Methods 0.000 description 6
- 230000009466 transformation Effects 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 5
- 238000000862 absorption spectrum Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000000835 fiber Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000011005 laboratory method Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000012925 reference material Substances 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 229920000742 Cotton Polymers 0.000 description 1
- 229920000995 Spectralon Polymers 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 229910052736 halogen Inorganic materials 0.000 description 1
- 150000002367 halogens Chemical class 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000010453 quartz Substances 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004579 scanning voltage microscopy Methods 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/28—Investigating the spectrum
- G01J3/2823—Imaging spectrometer
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N11/00—Investigating flow properties of materials, e.g. viscosity, plasticity; Analysing materials by determining flow properties
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
- G01N21/274—Calibration, base line adjustment, drift correction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/36—Textiles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N11/00—Investigating flow properties of materials, e.g. viscosity, plasticity; Analysing materials by determining flow properties
- G01N2011/006—Determining flow properties indirectly by measuring other parameters of the system
- G01N2011/008—Determining flow properties indirectly by measuring other parameters of the system optical properties
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/0058—Kind of property studied
- G01N2203/0092—Visco-elasticity, solidification, curing, cross-linking degree, vulcanisation or strength properties of semi-solid materials
- G01N2203/0094—Visco-elasticity
Abstract
According to an example aspect of the present invention, there is provided a method comprising determining at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of at least one calibration cellulose-based textile sample, retrieving a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample, determining a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity, estimating a viscosity of each calibration cellulose-based textile sample based on the mathematical model, evaluating the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose-based textile sample and determining, upon positive evaluation, an estimated viscosity of at least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
Description
ESTIMATING THE VISCOSITY OF TEXTILE MATERIALS
FIELD
[001] Embodiments of the present invention relate in general to estimating viscosity of textile materials.
BACKGROUND
[002] The demand for new textiles is constantly growing and the production of natural or virgin textile fibers cannot meet their demand. For example, production of cellulose and cotton fabrics from natural or virgin materials requires large amounts of natural resources and hence, recycling and reuse of waste textiles are becoming more and more important. Efficient recycling, however, requires sorting of waste textiles. Consequently, one challenge is that chemical recyclers would need to sort textiles based on viscosity to control the viscoelastic properties of textile fibers during recycling.
[003] Intrinsic viscosity may be used as an estimate of the viscoelastic properties of dissolved fibres, such as cellulose-based textile fibres, wherein said dissolving takes place during the production of man-made fibres. Intrinsic viscosity may be determined via chemical laboratory methods (e.g., SCAN-CM 15:88) and used to estimate the viscoelastic properties of dissolved textile fibres. These chemical laboratory methods are, however, slow, invasive and lead to the destruction of the samples. There is therefore a need to provide an improved method, apparatus and computer program for estimating the viscosity of textile materials, particularly for cellulose-based textile fibers.
SUMMARY OF THE INVENTION
[004] According to some aspects, there is provided the subject-matter of the independent claims. Some embodiments are defined in the dependent claims.
[005] According to a first aspect of the present invention, there is provided an apparatus comprising at least one processor; and at least one memory including computer
program code; the at least one memory and the computer program code being configured to, with the at least one processor, cause the apparatus at least to determine at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample, retrieve a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample, determine a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity, estimate a viscosity of each calibration cellulose-based textile sample based on the mathematical model, evaluate the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose-based textile sample and determine, upon positive evaluation, an estimated viscosity of least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
[006] Embodiments of the first aspect may comprise at least one feature from the following bulleted list:
• wherein the viscosities are intrinsic viscosities;
• wherein the at least one calibration cellulose-based textile sample and the at least one target cellulose-based textile sample are dry;
• wherein the mathematical model comprises weights, said weights being wavelength-specific;
• wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to transform the average spectrum of each hyperspectral image of each calibration cellulose-based textile sample to a value or a class that corresponds to the corresponding reference viscosity of each of the at least one calibration cellulose-based textile sample and estimate the viscosity of each of the at least one calibration cellulose-based textile sample based on the said value or the class;
• wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to: transform the average spectrum of each calibration cellulose-based textile sample to the value or the class that corresponds to the corresponding
reference viscosity of each of the at least one calibration cellulose-based textile sample by multiplying the average spectrum of each calibration cellulose-based textile sample with said weights;
• wherein a chemical composition of at least one of the at least one calibration cellulose-based textile sample is different compared to a chemical composition of the at least one target cellulose-based textile sample;
• An apparatus according to any of the preceding claims, wherein said hyperspectral images are near infrared images, such as with wavelengths from 450 nm to 2600 nm, preferably 1000 - 2500 nm;
• wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to receive, from a hyperspectral camera, said hyperspectral images of the at least one calibration cellulose-based textile sample and the at least one target cellulose-based textile sample;
• wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to calibrate the mathematical model using the average spectrum of said at least some pixels of each hyperspectral image and the corresponding reference viscosities of each of the at least one calibration cellulose-based textile sample, wherein said calibration comprises determining wavelength-specific weights, evaluation of said weights by comparing the reference viscosities with the viscosities or classes estimated by the model, removing upon negative evaluation samples that have large differences between the reference and estimated viscosities, determining other weights and evaluating the mathematical model with said other weights;
• wherein the mathematical model is a regression model and the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to determine a number of reliably estimated calibration cellulose-based textile samples, wherein a sample is reliably estimated when a difference between the estimated viscosity of the sample is within a threshold compared to a value of the reference viscosity of the sample and determine, if a weighted sum of said differences of a number of reliably estimated samples, or the number of
reliably estimated samples in general, is lower than a threshold, that said evaluation is positive, i.e., if the weighted sum or average of said differences of many samples is lower than the threshold;
• wherein the mathematical model is a classification model and the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to determine a number of correctly classified calibration cellulose-based textile samples, wherein a sample is classified correctly when the estimated viscosity of the sample belongs to a same class as the reference viscosity of the sample and determine, if the number, or a fraction, of correctly classified samples is larger than a threshold, that said evaluation is positive;
• An apparatus according to any of the preceding claims, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to remove at least one pixel from a background of each hyperspectral image to select a relevant region of interest of each hyperspectral image, calculate the average spectrum of each selected region of interest and determine the average spectrum of at least some pixels of each hyperspectral image as the average spectrum of the corresponding selected region of interest.
[007] According to a second aspect of the present invention, there is provided a method, comprising determining at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample, retrieving a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample, determining a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity, estimating a viscosity of each calibration cellulose-based textile sample based on the mathematical model, evaluating the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of the at least one calibration cellulose- based textile sample and determining, upon positive evaluation, an estimated viscosity of at least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
[008] According to a third aspect of the present invention, there is provided a computer program configured to perform the method.
BRIEF DESCRIPTION OF THE DRAWINGS
[009] FIGURE 1 illustrates operation of a hyperspectral imaging system in accordance with at least some embodiments of the present invention;
[0010] FIGURE 2 illustrates a hyperspectral image processing workflow in accordance with at least some embodiments of the present invention;
[0011] FIGURE 3 a illustrates choosing a region of interest in accordance with at least some embodiments of the present invention;
[0012] FIGURE 3b illustrates the effect of changing the size of the region of interest in accordance with at least some embodiments of the present invention;
[0013] FIGURE 4 illustrates an example apparatus capable of supporting at least some embodiments of the present invention;
[0014] FIGURE 5 illustrates a flow graph of a method in accordance with at least some embodiments of the present invention.
EMBODIMENTS
[0015] Embodiments of the present invention provide enhancements for estimating the viscosity of textile materials, in particular dry cellulose-based fabrics. Dry cellulose- based fabrics may refer for example to fabrics having a relative moisture content of less than 15%, preferably less than 10%, such as less than 5%.
[0016] More specifically, embodiments of the present invention provide reliable estimates of the average viscosity, like intrinsic viscosity, of cellulose-based textile samples, such as fabrics, without even touching the sample. The estimation can be done by exploiting hyperspectral imaging, to enable fast and fully automatized measurements, without wet chemical methods (e.g., SCAN-CM 15:58). Reliable estimates of viscosity enable controlling the raw material feed and the viscoelastic properties of dissolved cellulose-based
fibres during chemical textile recycling. For instance, there may be an optimal viscosity range and samples that belong to that range, i.e., class, may be selected for recycling.
[0017] According to some example embodiments of the present invention, hyperspectral near infrared images may be taken from a set of cellulose-based textile samples with known reference viscosities. The spectral fingerprints from said hyperspectral images may be used to develop a mathematical model, like a regression or classification model, to estimate the reference viscosities. The mathematical model may then be used to estimate the viscosity of a target cellulose-based textile sample based on a hyperspectral near infrared image of that sample.
[0018] FIGURE 1 illustrates operation of a hyperspectral imaging system in accordance with at least some embodiments of the present invention. Hyperspectral imaging system 100 may comprise hyperspectral camera 110, object 120 and apparatus 130. Hyperspectral camera 110 may be configured to take hyperspectral images of object 120. Object 120 may comprise one fabric at a time, comprising training, test or target cellulose- based textile samples. Said training and test samples may be referred to as calibration samples in general.
[0019] The chemical compositions of the calibration and target samples may be different, i.e., the calibration and target samples may be different. In some embodiments, a chemical composition of at least one of the at least one calibration cellulose-based textile sample may be different compared to a chemical composition of the at least one target cellulose-based textile sample. However, in some embodiments, at least one of the target samples may have the same, or similar, chemical composition as the at least one calibration sample, to enable control, and possibly continuous calibration as well.
[0020] Apparatus 130 may be a computer configured to provide an estimation of viscosity of object 120. Apparatus 130 may comprise for example at least one processor; and at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processor, cause the apparatus at least to estimate viscosity of object 120 from at least one image taken by hyperspectral camera 110.
[0021] In case of hyperspectral imaging, every image may comprise information of the near infrared region, like 450 nm - 2600 nm, preferably 1000 - 2500 nm. Near infrared
hyperspectral imaging may be used to analyse the chemical composition of materials, as the near infrared light and different materials interact characteristically.
[0022] The individual wavelength images may be stacked to form a hyperspectral data cube. Each hyperspectral data cube may contain spatial coordinates and a spectrum in each pixel. Hyperspectral imaging may thus combine digital imaging and spectroscopy.
[0023] The measured spectra may be used to identify materials based on the spectral fingerprints. Different materials may interact differently with electromagnetic radiation and thus have unique spectra. Measured spectra may be used to identify different areas or, for example, to identify material compounds in the sample.
[0024] Hyperspectral cameras, like camera 110, comprise an objective, a spectrograph and a grayscale camera. Hyperspectral camera 110 also includes at least one source of illumination, i.e., at least one light source 112. Angles of at least one light source 112 and at least one detector 114 need to be set optimally to reduce these effects and to obtain reliable imaging results. Diffuse illumination, detection and correct image preprocessing methods may be used to avoid these spectral and optical artefacts. Hyperspectral imaging may provide significant benefits for textile applications.
[0025] When light signal 122, produced by light source 112, interacts with material of object 120, absorbed light may be approximated to be the incident light subtracted with reflected light 124. Imaging applications usually comprise light source 112, object 120 that reflects the incident light and detector 114 that measures reflected light 124. Light source 112 may be directed to transmit light signal 122 to object 120 in a desired angle and reflected light 124, along with the reflection angle, may be measured using detector 114. Reflectance may be the fraction of incident light signal 122 that is reflected.
[0026] The collected raw hyperspectral images may be first acquired as detector signal intensity counts and the detector signal intensity counts may be transformed into reflectance values. Said transformation may be done using reflectance target and dark current measurements. Reflectance target spectra may be measured from standard reference materials, such as a Spectralon sheet, for which the wavelength-specific reflectance values may be known. Dark current may be obtained by measuring the spectra after blocking a lens of a camera, i.e., detector 114, so that light does not enter detector 114. Transformation of
the detector signal intensity counts to reflectance values may be calculated using the equation below
where R is the reflectance value, H is the sample spectra, D the dark current and W the reflectance target value. More complex calibration models may also be used. Accuracy of the calibration may be improved by measuring spectra from more than one standard reference material but that would make the calculation more complicated and timeconsuming. The reflectance transformations may be calculated on the individual pixel or average sample object levels.
[0027] After conducting desired calibrations, the reflectance values may be changed into absorbance units as follows
A = -log10(R), (2) where A is the absorbance unit and R contains the unitless reflectance values. Absorbance transformed spectra may show the absorbance band locations, which may be assigned to different substances.
[0028] The hyperspectral images may contain data of the sample and of the background. The sample pixels may be separated from the background for further analysis. In some embodiments, images of the calibration samples with known reference viscosities may be reflectance calibrated. The sample pixels in each sample image may be separated from the backgrounds using, for example, wavelength ratios or multivariate methods such as Principal Component Analysis, PCA, and averaged to provide one average spectrum per sample. The average spectra of the calibration samples may then be transformed into pseudoabsorbance and combined into a calibration set. These calibration set spectra may then be preprocessed and a mathematical model developed based on the known reference intrinsic viscosities. The viscosities of unknown, target samples may then be estimated by taking new hyperspectral images of the target samples and using the developed mathematical model on the new hyperspectral images.
[0029] Preprocessing may be used to prepare the reflectance/absorbance signals for modelling by highlighting important changes in the spectra and reducing the unimportant
ones due to, e.g., uneven lighting. For instance, Standard Normal Variate, SNV, Multiplicative Scatter Correction, MSC, and/or different derivative methods may be used for the preprocessing.
[0030] Once the signals have been preprocessed, a mathematical model, such as a regression or a classification model may be built. Regression could be important as the lab analysis for viscosity may provide continuous numerical values essentially making this a regression problem. Example regression methods comprise at least Multiple Linear Regression, MLR, Partial Least Squares, PLS, Support Vector Machines, SVM, and neural networks. Example classification methods comprise Linear Discriminant Analysis, LDA, Quadratic Discriminant Analysis, QDA, PLS- Discriminant Analysis, PLS-DA, Support Vector Domain Description, SVDD. Viscosity of a target sample may then be estimated using the mathematical model.
[0031] LIGURE 2 illustrates a hyperspectral image processing workflow in accordance with at least some embodiments of the present invention. At step 210, imaging of hyperspectral images may be performed by hyperspectral camera 110. Lor instance, the images may be scanned in line-scanning mode with a defined number of spatial pixels in one line and spectral wavelength bands for each pixel. Spectral bands may cover the wavelength region 1000-2500 nm with a given spectral sampling width and spectral resolution. Samples may be illuminated for example with polychromatic light produced by quartz halogen lamps. Each sample may be imaged separately with a reflectance target.
[0032] The field of view of hyperspectral camera 110 may be set to cover the textile samples. During imaging the samples may be moved under detector 114, a constant distance away from the objective of detector 114. The speed of a moving table, i.e., the location of object 120, may be set based on the frame rate and integration time of the camera 110.
[0033] Loading of images to apparatus 130 may be performed as well. That is, hyperspectral camera 110 may transmit said images to apparatus 130, possibly via a communication network. Said hyperspectral images may be raw hyperspectral images, measured in detector signal intensity counts. Apparatus 130 may then determine, at step 210, at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of at least one calibration cellulose-based textile sample.
[0034] At step 220, apparatus 130 may retrieve a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample, for example from its memory or from another apparatus, such as a server.
[0035] At step 230, apparatus 130 may determine a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity. That is, the mathematical model may create a link between one determined average spectrum and one reference viscosity, and a link between another determined average spectrum and another reference viscosity. Said links may be used together to form an estimated general link, which may be further exploited to associate any determined average spectrum and a corresponding, estimated viscosity.
[0036] The mathematical model may comprise weights. Said weights may be wavelength-specific and may be determined by, e.g., projecting calibration spectra onto latent variables to capture the covariance between by the spectra and reference viscosities or to maximize the ratio of between-class and within-class differences by linear or non-linear methods.
[0037] In some embodiments, apparatus 130 may transform the average spectrum of each hyperspectral image of each calibration cellulose-based textile sample to a value or a class that corresponds to the corresponding reference viscosity of the at least one calibration cellulose-based textile sample and estimate the viscosity of each of the at least one calibration cellulose-based textile sample based on the value or the class. For instance, if the corresponding reference viscosity of one sample belongs to a class “Optimal viscosity, 350 - 550 ml/g”, the determined average spectrum of the sample may be associated with that class. As this is done for multiple samples, the viscosity of the sample may be estimated such that it belongs to the same class, if the mathematical model works well enough. Similarly, if the corresponding reference viscosity of the sample has a value of 450 ml/g, the determined average spectrum may be associated with values 450 ml/g ± 25 ml/g, wherein 25 ml/g is a threshold. The viscosity of each hyperspectral image of the sample may be estimated to have a value of 450 ml/g ± 25 ml/g, if the mathematical model works well enough.
[0038] Apparatus 130 may further transform the average spectrum of each calibration cellulose-based textile sample to the value or the class that corresponds to the viscosity of each of the at least one calibration cellulose-based textile sample by multiplying the average
spectrum of each calibration cellulose-based textile sample with said weights. That is, the mathematical model may comprise weights and thus, the transformation may depend on weighting as well. The purpose is to find such weights that give the correct value and/or class for all the samples and the mathematical model may be then evaluated by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose-based textile sample, i.e., by checking whether an estimated viscosity of each sample has about the same value (within the threshold) or is in the same class as the corresponding reference viscosity. If there are large differences between the reference and estimated viscosities (when a weighted sum of said differences of a number of samples is larger than the threshold or a number of incorrectly classified samples is larger than a threshold, wherein the threshold may be for example 5% or 10%), samples with large differences may be removed and other weights determined after that based on the remaining samples. The mathematical model may be evaluated again after that using said other samples.
[0039] At step 240, apparatus 130 may estimate a viscosity of each calibration cellulose-based textile sample based on the mathematical model. That is, the estimated general link may be exploited to associate the determined average spectrum of each calibration cellulose-based textile sample with a viscosity value or a class.
[0040] At step 250, apparatus 130 may evaluate the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of the at least one calibration cellulose-based textile sample. Thus, the accuracy of the mathematical model, and the estimated general link, may be evaluated by comparing the estimated viscosities values to the reference viscosity values, to find out differences between the estimated viscosities values and the reference viscosity values and a weighted sum or average of said differences that is within a threshold. The threshold may be for example 5, 15, 25 ml/g or 25, 225, 625 (ml/g)2.
[0041] Alternatively, or in addition, the accuracy of the mathematical model, and the estimated general link, may be evaluated by comparing the estimated viscosities classes to the reference viscosity classes, to find out a number of estimated viscosities that are classified into the same class as the corresponding reference viscosity.
[0042] For instance, if the mathematical model is a regression model, apparatus 130 may determine a number of reliably estimated calibration cellulose-based textile samples,
wherein a sample is reliably estimated when the difference between the estimated viscosity of the sample and the corresponding reference viscosity is within a threshold and, if the weighted sum or average of said differences across many samples is lower than a threshold (e.g., 25 ml/g or 625 ml2/g2), further determine that said evaluation is positive.
[0043] Alternatively, or in addition, if the mathematical model is a classification model, apparatus 130 may determine a number of correctly classified calibration cellulose- based textile samples, wherein a sample is classified correctly when the estimated viscosity of the sample belongs to a same class as the reference viscosity of the sample and if the number of correctly classified samples is larger than a threshold (e.g., 90% or 95%), determine that said evaluation is positive.
[0044] In case of determining that said evaluation is negative, the mathematical model may be further calibrated. Said calibration may comprise determining wavelength-specific weights, evaluation of said weights by comparing the reference viscosities with the viscosities or classes estimated by the model, removing upon negative evaluation samples that have large differences between the reference and estimated viscosities, determining other weights and evaluating the mathematical model again with said other weights.
[0045] For instance, weights/ coefficients may be determined for a group of calibration samples what have deemed representative of the range wherein it is desirable to make the regression/classification model to function reliably. Once these weights have been determined, the estimated viscosity or class values may be determined and the measured vs. estimated viscosities may be compared. The complexity and/or accuracy of the model may be evaluated to, e.g., decrease the number of latent variables and identify potential outliers, i.e., samples that show large differences that due to some reason, and could potentially be removed from the calibration set for the model to function better. This outlier detection may also be done based on some residual distances (different distance metrics exist in multivariate space depending on the model), as it is normally easier to see the discrepancies. This would normally require some iterative steps in potentially removing samples and redetermining the weights/coefficients, and complexity/accuracy. Once this is done, the most important wavelengths may be selected to make the model simpler and more reliable. Finally, a test set may be used to determine model performance based on samples which were not used for calibrating the model.
[0046] At step 260, apparatus 130 may determine, upon positive evaluation, an estimated viscosity of least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model. That is, the estimated general link may be exploited to associate the determined average spectrum of each target cellulose-based textile sample with a viscosity value or a class.
[0047] In some example embodiments, apparatus 130 may remove at least one pixel from a background of each hyperspectral image to select a relevant region of interest of each hyperspectral image, calculate the average spectrum of each selected region of interest and determine the average spectrum of at least some pixels of each hyperspectral image as the average spectrum of the corresponding selected region of interest.
[0048] For instance, said hyperspectral images may be transformed into reflectance images possibly after background removal. Reflectance target and dark current measurements may be used to transform the raw signal values into reflectance using for example Eq. (1). The reflectance transformed images may still be large hyperspectral data cubes also including information of the background.
[0049] The sample pixels may be separated from the background pixels using, for example, wavelength ratios or PCA. As a result, only the sample pixels may be chosen for further analysis and all sample pixels may be used to calculate the average spectrum for each sample.
[0050] After that, transformation to absorbance may be performed. The absorbance transformation shown in Eq. (2) may be calculated for the individual image pixels or the average sample signal.
[0051] Calculation of the average sample spectra may be performed and a classification model may be determined using the preprocessed average absorbance spectrum of the sample pixels. Table 1 presents an example of a classification model.
Table 1 An example of class definitions for a classification model
[0052] Test set sample spectra of at least one test cellulose-based textile sample, may be then used to determine the accuracy of the final regression or classification model. The test set sample spectra may be preprocessed with the same methods as the samples in the training set. Apparatus 130 may evaluate the accuracy of the regression or classification model using the measured and estimated viscosity values for at least one test cellulose-based textile sample. Alternatively, the accuracy of the regression or classification model may be determined using cross-validation. For instance, if there are no test set samples available, the accuracy or reliability of the regression or classification model could be estimated using cross-validation.
[0053] The classification model may be evaluated using a confusion matrix for example, as shown in Table 2 below, wherein reference viscosity of the at least one test cellulose-based textile sample are shown in columns while viscosity estimated from said hyperspectral images of the at least one test cellulose-based textile sample are shown in rows. In this example, the rate of correctly estimated samples is 94.7%.
Table 2 Confusion matrix of the test set classification results
[0054] In some example embodiments, apparatus 130 may determine the mathematical model using the average spectrum of said at least some pixels of each hyperspectral image and the corresponding reference viscosities of each of the at least one calibration cellulose-based textile sample. Said mathematical model may comprise of wavelength-specific weights and the determination of correct weights may be evaluated by comparing the reference viscosities with the viscosities or classes predicted by the model. Said mathematical model may be further examined by identifying and removing upon
negative evaluation samples that have large differences between the estimated and reference viscosities, determining other weights and evaluating the mathematical model with said other weights. Said mathematical model may also be determined by choosing a number of wavelengths or spectral bands which is lower than the number of wavelengths or spectral bands measured by the camera detector.
[0055] FIGURE 3 a illustrates choosing a region of interest in accordance with at least some embodiments of the present invention. Classification results may be calculated using average absorbance spectra of the samples in the calibration, cross-validation or test set. All of the sample pixels may be used to calculate the average spectra or some region of interest may be chosen.
[0056] In FIGURE 3a is shown an image of a sample and the 200x200 pixel region of interest that may be used to calculate the average absorbance spectrum. Using comparable regions of interest for all samples instead of all the sample pixels to calculate the average spectra may improve further analysis of the samples and corresponding reference viscosities.
[0057] FIGURE 3b illustrates the effect of changing the size of the region of interest expressed as the number of hyperspectral image pixels on the preprocessed average absorbance spectra in accordance with at least some embodiments of the present invention.
[0058] FIGURE 4 illustrates an example apparatus capable of supporting at least some embodiments of the present invention. Illustrated is apparatus 400, which may comprise or correspond to apparatus 130 of FIGURE 1.
[0059] Comprised in apparatus 400 may be processing unit, i.e., processing element, 410, which may further comprise, for example, a single- or multi-core processor wherein a single-core processor comprises one processing core and a multi-core processor comprises more than one processing core. Processing unit 410 may comprise, in general, a control device. Processing unit 410 may comprise one or more processors. Processing unit 410 may be a control device. Processing unit 410 may comprise at least one Application-Specific Integrated Circuit, ASIC. Processing unit 410 may comprise at least one Field- Programmable Gate Array, FPGA. Processing unit 410 may be means for performing method steps in apparatus 400. Processing unit 410 may be configured, at least in part by computer instructions, to perform actions.
[0060] Apparatus 400 may comprise memory 420. Memory 420 may comprise Random- Access Memory, RAM, and/or permanent memory. Memory 420 may comprise at least one RAM chip. Memory 420 may comprise solid-state, magnetic, optical and/or holographic memory, for example. Memory 420 may be at least in part accessible to processing unit 410. Memory 420 may be at least in part comprised in processing unit 410. Memory 420 may be means for storing information, such as a phase and amplitude of a reflected signal. Memory 420 may comprise computer instructions that processing unit 410 is configured to execute. When computer instructions configured to cause processing unit 410 to perform certain actions are stored in memory 420, and apparatus 400 overall is configured to run under the direction of processing unit 410 using computer instructions from memory 420, processing unit 410 and/or its at least one processing core may be configured to perform said certain actions. Memory 420 may be at least in part comprised in processing unit 410. Memory 420 may be at least in part external to apparatus 400 but accessible to apparatus 400.
[0061] Apparatus 400 may comprise a transmitter 430. The transmitter 430 may be used to transmit via a communication network, for example, to transmit to hyperspectral camera 110. Apparatus 400 may also comprise a receiver 440. The received 440 may be used to receive via a communication network, for example, to receive from hyperspectral camera 110.
[0062] Apparatus 400 may also comprise a user interface, UI, 450. UI 450 may comprise at least a display or a touchscreen. A user may be able to operate apparatus 400 via UI 450. Also, UI 450 may be used for displaying information to the user. For example, UI 450 may be used for displaying viscosity values estimated from hyperspectral images and take as an input reference viscosity values of samples.
[0063] Processing unit 410 may be furnished with a transmitter arranged to output information from processing unit 410, via electrical leads internal to apparatus 400, to other devices comprised in apparatus 400. Such a transmitter may comprise a serial bus transmitter arranged to, for example, output information via at least one electrical lead to memory 420 for storage therein. Alternatively to a serial bus, the transmitter may comprise a parallel bus transmitter. Likewise processing unit 410 may comprise a receiver arranged to receive information in processing unit 410, via electrical leads internal to apparatus 400, from other devices comprised in apparatus 400. Such a receiver may comprise a serial bus receiver
arranged to, for example, receive information via at least one electrical lead from receiver 440 for processing in processing unit 410. Alternatively to a serial bus, the receiver may comprise a parallel bus receiver.
[0064] Processing unit 410, memory 420, transmitter 430, receiver 440 and/or UI 450 may be interconnected by electrical leads internal to apparatus 400 in a multitude of different ways. For example, each of the aforementioned devices may be separately connected to a master bus internal to apparatus 400, to allow for the devices to exchange information. However, as the skilled person will appreciate, this is only one example and depending on the embodiment various ways of interconnecting at least two of the aforementioned devices may be selected without departing from the scope of the present invention
[0065] FIGURE 5 illustrates a flow graph of a method in accordance with at least some embodiments of the present invention. The phases of the illustrated method may be performed by apparatus 130, or by a control device configured to control the functioning thereof, possibly when installed therein. Also, there may be a computer program configured to perform the illustrated method.
[0066] The method may comprise, at step 510, determining at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample. The method may also comprise, at step 520, retrieving a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample. At step 530, the method may comprise determining a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity. At step 540, the method may comprise estimating a viscosity of each calibration cellulose- based textile sample based on the mathematical model. At step 550, the method may comprise evaluating the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of the at least one calibration cellulose-based textile sample. Finally, at step 560, the method may comprise determining, upon positive evaluation, an estimated viscosity of least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the model.
[0067] It is to be understood that the embodiments of the invention disclosed are not limited to the particular structures, process steps, or materials disclosed herein, but are
extended to equivalents thereof as would be recognized by those ordinarily skilled in the relevant arts. It should also be understood that terminology employed herein is used for the purpose of describing particular embodiments only and is not intended to be limiting.
[0068] Reference throughout this specification to one embodiment or an embodiment means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Where reference is made to a numerical value using a term such as, for example, about or substantially, the exact numerical value is also disclosed.
[0069] As used herein, a plurality of items, structural elements, compositional elements, and/or materials may be presented in a common list for convenience. However, these lists should be construed as though each member of the list is individually identified as a separate and unique member. Thus, no individual member of such list should be construed as a de facto equivalent of any other member of the same list solely based on their presentation in a common group without indications to the contrary. In addition, various embodiments and example of the present invention may be referred to herein along with alternatives for the various components thereof. It is understood that such embodiments, examples, and alternatives are not to be construed as de facto equivalents of one another, but are to be considered as separate and autonomous representations of the present invention.
[0070] In an exemplary embodiment, an apparatus, such as, for example, apparatus 130, may comprise means for carrying out the embodiments described above and any combination thereof.
[0071] In an exemplary embodiment, a computer program may be configured to cause a method in accordance with the embodiments described above and any combination thereof. In an exemplary embodiment, a computer program product, embodied on a non-transitory computer readable medium, may be configured to control a processing unit to perform a process comprising the embodiments described above and any combination thereof.
[0072] In an exemplary embodiment, an apparatus, such as, for example, apparatus 130, may comprise at least one processing unit, and at least one memory including computer program code, wherein the at least one memory and the computer program code are
configured to, with the at least one processing unit, cause the apparatus at least to perform the embodiments described above and any combination thereof.
[0073] Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the preceding description, numerous specific details are provided, such as examples of lengths, widths, shapes, etc., to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.
[0074] While the forgoing examples are illustrative of the principles of the present invention in one or more particular applications, it will be apparent to those of ordinary skill in the art that numerous modifications in form, usage and details of implementation can be made without the exercise of inventive faculty, and without departing from the principles and concepts of the invention. Accordingly, it is not intended that the invention be limited, except as by the claims set forth below.
[0075] The verbs “to comprise” and “to include” are used in this document as open limitations that neither exclude nor require the existence of also un-recited features. The features recited in depending claims are mutually freely combinable unless otherwise explicitly stated. Furthermore, it is to be understood that the use of "a" or "an", that is, a singular form, throughout this document does not exclude a plurality.
INDUSTRIAL APPLICABILITY
[0076] At least some embodiments of the present invention find industrial application in recycling and reusing waste textiles.
ACRONYMS LIST
ASIC Application-Specific Integrated Circuit
CVA Canonical Variate Analysis
FPGA Field-Programmable Gate Array
LDA Linear Discriminant Analysis
MLR Multiple Linear Regression
MSC Multiplicative Scatter Correction PCA Principal Components Analysis
PLS Partial Least Squares
PLS-DA PLS- Discriminant Analysis
QDA Quadratic Discriminant Analysis
RAM Random- Access Memory SNV Standard Normal Variate
SVD Singular Value Decomposition
SVDD Support Vector Domain Description
SVM Support Vector Machines
UI User Interface
REFERENCE SIGNS LIST
Claims
1. An apparatus comprising at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code being configured to, with the at least one processor, cause the apparatus at least to:
- determine at least one hyperspectral image of at least one calibration cellulose-based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample;
- retrieve a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample;
- determine a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity;
- estimate a viscosity of each calibration cellulose-based textile sample based on the mathematical model;
- evaluate the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose- based textile sample; and
- determine, upon positive evaluation, an estimated viscosity of at least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
2. An apparatus according to claim 1, wherein the viscosities are intrinsic viscosities.
3. An apparatus according to claim 1 or claim 2, wherein the at least one calibration cellulose-based textile sample and the at least one target cellulose-based textile sample are dry.
4. An apparatus according to any of the preceding claims, wherein the mathematical model comprises weights, said weights being wavelength-specific.
5. An apparatus according to claim 4, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- transform the average spectrum of each hyperspectral image of each calibration cellulose-based textile sample to a value or a class that corresponds to the corresponding reference viscosity of each of the at least one calibration cellulose- based textile sample; and
- estimate the viscosity of each of the at least one calibration cellulose-based textile sample based on the value or the class.
6. An apparatus according to claim 5, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- transform the average spectrum of each calibration cellulose-based textile sample to the value or the class that corresponds to the corresponding reference viscosity of each of the at least one calibration cellulose-based textile sample by multiplying the average spectrum of each calibration cellulose-based textile sample with said weights.
7. An apparatus according to any of the preceding claims, wherein a chemical composition of at least one of the at least one calibration cellulose-based textile sample is different compared to a chemical composition of the at least one target cellulose-based textile sample.
8. An apparatus according to any of the preceding claims, wherein said hyperspectral images are near infrared images, such as with wavelengths from 450 nm to 2600 nm, preferably 1000 - 2500 nm.
9. An apparatus according to any of the preceding claims, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- receive, from a hyperspectral camera, said hyperspectral images of the at least one calibration cellulose-based textile sample and the at least one target cellulose-based textile sample.
10. An apparatus according to any of the preceding claims, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- calibrate the mathematical model using the average spectrum of said at least some pixels of each hyperspectral image and the corresponding reference viscosities of each of the at least one cellulose-based textile sample, wherein said calibration comprises determining wavelength-specific weights, evaluation of said weights by comparing the reference viscosities with the viscosities or classes estimated by the model, removing upon negative evaluation samples that have large differences between the reference and estimated viscosities, determining other weights and evaluating the mathematical model with said other weights.
11. An apparatus according to any of the preceding claims, wherein the mathematical model is a regression model and the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- determine a number of reliably estimated calibration cellulose-based textile samples, wherein a sample is reliably estimated when a difference between the estimated viscosity of the sample is within a threshold compared to a value of the reference viscosity of the sample; and
- determine, if a weighted sum or average of said differences of many samples is lower than a threshold, that said evaluation is positive.
12. An apparatus according to any of claims 1 to 10, wherein the mathematical model is a classification model and the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- determine a number of correctly classified calibration cellulose-based textile samples, wherein a sample is classified correctly when the estimated viscosity of the sample belongs to a same class as the reference viscosity of the sample; and
- determine, if the number of correctly classified samples is larger than a threshold, that said evaluation is positive.
13. An apparatus according to any of the preceding claims, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to:
- remove at least one pixel from a background of each hyperspectral image to select a relevant region of interest of each hyperspectral image;
- calculate the average spectrum of each selected region of interest; and
- determine the average spectrum of at least some pixels of each hyperspectral image as the average spectrum of the corresponding selected region of interest.
14. A method comprising:
- determining at least one hyperspectral image of at least one calibration cellulose- based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample;
- retrieving a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample;
- determining a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity;
- estimating a viscosity of each calibration cellulose-based textile sample based on the mathematical model;
- evaluating the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose- based textile sample; and
- determining, upon positive evaluation, an estimated viscosity of at least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
15. A computer program configured to perform:
- determining at least one hyperspectral image of at least one calibration cellulose- based textile sample and an average spectrum of at least some pixels of each hyperspectral image of each of the at least one calibration cellulose-based textile sample;
- retrieving a corresponding reference viscosity for each of the at least one calibration cellulose-based textile sample;
- determining a mathematical model by associating the determined average spectrum of each hyperspectral image of each of the at least one calibration cellulose-based textile sample to the corresponding reference viscosity;
- estimating a viscosity of each calibration cellulose-based textile sample based on the mathematical model;
- evaluating the mathematical model by comparing each estimated viscosity to the corresponding reference viscosity of each of the at least one calibration cellulose- based textile sample; and
- determining, upon positive evaluation, an estimated viscosity of at least one target cellulose-based textile sample from a hyperspectral image of the at least one target cellulose-based textile sample using the mathematical model.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FI20225430A FI20225430A1 (en) | 2022-05-16 | 2022-05-16 | Estimating the viscosity of textile materials |
| FI20225430 | 2022-05-16 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023222945A1 true WO2023222945A1 (en) | 2023-11-23 |
Family
ID=86468945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/FI2023/050264 WO2023222945A1 (en) | 2022-05-16 | 2023-05-12 | Estimating the viscosity of textile materials |
Country Status (2)
| Country | Link |
|---|---|
| FI (1) | FI20225430A1 (en) |
| WO (1) | WO2023222945A1 (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999067722A1 (en) * | 1998-06-22 | 1999-12-29 | Umetri Aktiebolag | Method and arrangement for calibration of input data |
| MD3325F1 (en) * | 2006-07-17 | 2007-05-31 | Universitatea De Stat Din Moldova | Method of dating the old cellulose textile materials |
| CN109142153A (en) * | 2018-11-22 | 2019-01-04 | 北方天普化工有限公司 | A kind of cotton fiber Viscosity Analysis method |
| US20190218362A1 (en) * | 2018-01-12 | 2019-07-18 | Tyton Biosciences, Llc | Methods for recycling cotton and polyester fibers from waste textiles |
-
2022
- 2022-05-16 FI FI20225430A patent/FI20225430A1/en unknown
-
2023
- 2023-05-12 WO PCT/FI2023/050264 patent/WO2023222945A1/en unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999067722A1 (en) * | 1998-06-22 | 1999-12-29 | Umetri Aktiebolag | Method and arrangement for calibration of input data |
| MD3325F1 (en) * | 2006-07-17 | 2007-05-31 | Universitatea De Stat Din Moldova | Method of dating the old cellulose textile materials |
| US20190218362A1 (en) * | 2018-01-12 | 2019-07-18 | Tyton Biosciences, Llc | Methods for recycling cotton and polyester fibers from waste textiles |
| CN109142153A (en) * | 2018-11-22 | 2019-01-04 | 北方天普化工有限公司 | A kind of cotton fiber Viscosity Analysis method |
Non-Patent Citations (5)
| Title |
|---|
| ANONYMOUS: "Hyperspectral imaging to estimate intrinsic viscosity of cellulose-based fabrics - Trepo", 16 May 2022 (2022-05-16), XP093056394, Retrieved from the Internet <URL:https://trepo.tuni.fi/handle/10024/140203?show=full> [retrieved on 20230621] * |
| LINDA CSÉFALVAYOVÁ; ET AL: "Quantitative NIR Chemical Imaging in Heritage Science", ANALYTICAL CHEMISTRY, vol. 83, no. 13, 23 March 2011 (2011-03-23), US, pages 5101 - 5106, XP055669854, ISSN: 0003-2700, DOI: 10.1021/ac200986p * |
| MAHGOUB HEND ET AL: "Analytical robustness of quantitative NIR chemical imaging for Islamic paper characterization", PROCEEDINGS OF SPIE; [PROCEEDINGS OF SPIE ISSN 0277-786X VOLUME 10524], SPIE, US, vol. 10331, 11 July 2017 (2017-07-11), pages 103310P - 103310P, XP060092399, ISBN: 978-1-5106-1533-5, DOI: 10.1117/12.2271971 * |
| MAHLAMÄKI ELLA: "HYPERSPECTRAL IMAGING TO ESTIMATE INTRINSIC VISCOSITY OF CELLULOSE-BASED FABRICS", 16 May 2022 (2022-05-16), XP093056404, Retrieved from the Internet <URL:https://trepo.tuni.fi/bitstream/handle/10024/140203/Mahlam%C3%A4kiElla.pdf?sequence=2&isAllowed=y> [retrieved on 20230621] * |
| MÄKELÄ MIKKO ET AL: "Hyperspectral near infrared image calibration and regression", ANALYTICA CHIMICA ACTA, ELSEVIER, AMSTERDAM, NL, vol. 1105, 10 January 2020 (2020-01-10), pages 56 - 63, XP086072910, ISSN: 0003-2670, [retrieved on 20200110], DOI: 10.1016/J.ACA.2020.01.019 * |
Also Published As
| Publication number | Publication date |
|---|---|
| FI20225430A1 (en) | 2023-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Grabowski et al. | Automatic pigment identification from hyperspectral data | |
| Tatzer et al. | Industrial application for inline material sorting using hyperspectral imaging in the NIR range | |
| JP5496509B2 (en) | System, method, and apparatus for image processing for color classification and skin color detection | |
| CN110717368A (en) | Qualitative classification method for textiles | |
| US20080266430A1 (en) | Method and system for optimizing an image for improved analysis of material and illumination image features | |
| Dorrepaal et al. | Tutorial: Time series hyperspectral image analysis | |
| Brusa et al. | Increasing the precision of canopy closure estimates from hemispherical photography: Blue channel analysis and under-exposure | |
| US9606056B2 (en) | Selection of spectral bands or filters for material classification under multiplexed illumination | |
| CN110363125A (en) | Using the method for Model Transfer identification different cultivars Citrus Huanglongbing pathogen | |
| Yang et al. | Effect of smart phone cameras on color-based prediction of soil organic matter content | |
| Mäkelä et al. | Hyperspectral near infrared image calibration and regression | |
| Li et al. | A novel hyperspectral imaging and modeling method for the component identification of woven fabrics | |
| Guo et al. | Detection of foreign materials on surface of ginned cotton by hyper-spectral imaging | |
| Benoit et al. | On the value of the Kullback–Leibler divergence for cost-effective spectral imaging of plants by optimal selection of wavebands | |
| Sun et al. | Surface gloss evaluation of apples based on computer vision and support vector machine method | |
| WO2021195817A1 (en) | Method for extracting spectral information of object to be detected | |
| Gao et al. | Precautionary analysis of sprouting potato eyes using hyperspectral imaging technology | |
| WO2023222945A1 (en) | Estimating the viscosity of textile materials | |
| US20230162354A1 (en) | Artificial intelligence-based hyperspectrally resolved detection of anomalous cells | |
| CN114184599B (en) | Single-cell Raman spectrum acquisition number estimation method, data processing method and device | |
| CN113887543B (en) | Luggage counterfeit discrimination method based on hyperspectral characteristics and spectrum acquisition device | |
| CN110335249A (en) | Citrus Huanglongbing pathogen detection method based on high light spectrum image-forming technology | |
| CN115424006A (en) | Multi-source multi-level data fusion method applied to crop phenotypic parameter inversion | |
| Mahgoub et al. | Quantitative chemical near-infrared hyperspectral imaging of Islamic paper | |
| Liu et al. | Multispectral LiDAR point cloud highlight removal based on color information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23725369 Country of ref document: EP Kind code of ref document: A1 |