Optical measurement method of beard curve of colored cotton fibers based on approximate secondary reflectivity and bulk correction

Abstract

The present invention discloses an optical measurement method for colored cotton fiber whisker tuft curves based on approximate secondary reflectivity and bulk correction. The method comprises the following steps: firstly, obtaining a fiber reflection signal and a fiber whisker transmission signal by using red, yellow and blue three-channel monochromatic light; then correcting the reflectivity of the colored cotton fiber with a fixed stacking density by using a secondary estimation optical algorithm for the monochromatic light reflectivity; then, calculating an optical linear density distribution curve of a random double-ended whisker tuft by using the light transmission signal obtained from the whisker tuft image, the corrected reflectivity signal and an optical surface density algorithm; finally, calculating a colored cotton fiber whisker tuft curve by using an optical correction algorithm for the bulkiness of combed cotton fibers or lint loose fibers. The method is applicable to measuring the lengths of cotton fibers of different colors and is more accurate than a method directly using grayscale reflectivity or ideal reflectivity. The method can also be used to measure the mass density distribution of colored cotton fibers or polymer stacking materials.

Description

Technical field

[0001] The invention relates to the technical field of an optical measurement method for a beard tuft curve of colored cotton fibers, in particular to the technical field of an optical measurement method for a beard tuft curve of colored cotton fibers based on approximate secondary reflectivity and bulk correction. [Background Technology]

[0002] Colored cotton generally includes dyed cotton and naturally colored cotton. Colored cotton fiber length is a key physical characteristic parameter, influencing fiber quality, yarn spinnability, yarn quality, and genetic modification and breeding of naturally colored cotton. It also plays a crucial role in determining price and usage. However, measuring colored cotton fiber length remains difficult.

[0003] The two main international standard automated cotton fiber length measuring instruments currently available are the HVI (High Volume Instrument) and the AFIS (Advanced Fiber Information System). The HVI was developed by Spinlab in the 1970s based on Heretl's photogrammetric curve theory. It has become the international standard for measuring white cotton fiber length, but it cannot be used to measure short fiber content or colored fibers. The AFIS uses infrared light and airflow to measure the length of at least 3,000 single cotton fibers, providing an indicator of cotton fiber length. However, the sample preparation process can cause fiber breakage, resulting in only a portion of the fibers being measured (9-33%). Furthermore, the direction and curvature of the fibers in the airflow can lead to an underestimate of fiber length.

[0004] In addition, Wang Fumei and Wu Hongyan proposed a fast and low-cost fiber length measurement method in 2012 (Announcement No. CN102853775B), which includes sample preparation, acquisition of whisker clump transmission images and calculation of whisker clump linear density curves. The fiber length index can be calculated based on the whisker clump linear density curve. However, this method calculates the thickness information of the whisker clump based on the Lambert-Beer law, while the LB law only considers the absorption of light and ignores light scattering and reflection, resulting in deviations in the final result. In order to improve the accuracy of optical signal analysis in this method, Wu Meiqin et al. derived an algorithm in 2015 to calculate the relative surface density at any point of a layered fiber assembly from the transmitted and reflected optical signals of infinitely thick materials (Announcement No. CN105403482B), and named it the Wu-Wang algorithm. The optical density obtained by this algorithm has a high consistency with the true density, and its accuracy is higher than that of the Lambert-Beer law. It is applicable to a variety of fibers such as cotton, wool and kapok.

[0005] However, the above methods are only for the measurement of white cotton and wool fibers. That is to say, the above measurement method cannot directly use the reflection image value to measure the length index of colored fibers. In 2020, Heng Chong (Digital Cashmere Color Measurement and Its Application in Length Measurement, Heng Chong et al., Journal of Textile Research, Vol. 41, No. 12, December 2020) constructed the brightness value L converted from R, G, and B values ​​and compared it with the ideal R value of the grayscale channel measured by Datacolor spectrophotometer. ∞ The conversion model between the two was applied to the cashmere fiber length test, and a light purple cashmere beard curve was obtained that was highly consistent with the result of the hand arrangement method. However, the ideal R ∞ Failure to consider factors such as iris and noise along the light path, coupled with the narrow color gamut of natural cashmere and the wavelength dependence of fluorescent light sources far exceeding that of monochromatic light, results in inaccurate curves for colored cotton fiber beards. Therefore, developing appropriate optical measurement methods is necessary for measuring the length of cotton fiber samples with diverse colors and a wide variety of varieties. [Summary of the invention]

[0006] The purpose of the present invention is to solve the problems in the prior art and propose an optical measurement method for the whisker tuft curve of colored cotton fibers. The method comprises a second-order approximation correction method that takes into account influencing factors such as optical iris and noise on the propagation path when measuring the reflectivity of an infinitely thick colored fiber aggregate using monochromatic light, and an optical bulk correction algorithm on both sides of the whisker tuft curve clamping area. The method can be applied to measure the length of cotton fibers of different colors and is more accurate than methods that directly use grayscale reflectivity or ideal reflectivity. The method can also be used to measure the mass density distribution of colored cotton fibers or polymer stacking materials.

[0007] To achieve the above object, the present invention proposes an optical measurement method for colored cotton fiber beard tuft curve based on approximate secondary reflectivity and bulk correction, comprising the following steps:

[0008] Step 1: Image measurement and acquisition:

[0009] Obtain a layered material sample composed of colored cotton fibers or other colored polymer objects stacked to infinite thickness and measure its reflection image under uniform red, green, and blue (R, G, B) monochromatic light illumination. Then, use the colored cotton fibers to make a double-ended random whisker and measure its transmission image under uniform red, green, and blue (R, G, B) monochromatic light illumination. The layered material sample must ensure that the stacking density of the fibers is fixed and parallel during reflection measurement, while the double-ended random whisker of the colored cotton fibers must ensure that the fibers are parallel during transmission measurement.

[0010] Step 2: Correction of optical algorithm for secondary estimation of monochromatic light reflectance:

[0011] The instrumental reflectance under the red, green, and blue (R, G, B) channels is calculated using formula (1):

[0012] r i =U i / 255 (1);

[0013] Where U i are the values ​​of the R, G, and B channels obtained from the reflection image measured and collected in step 1, i = R, G, B;

[0014] The ideal reflectivity is obtained from the relationship formula between the imager and each sensor:

[0015] U i =∫p(λ)e(λ)f(λ)s i (λ)dλ 1≤i≤3 (2);

[0016] Where p(λ) is the spectral transmission efficiency of the light path, e(λ) is the spectral density distribution of the light source, and f(λ) is the spectral reflectance of the sample;

[0017] The sensitivity curve of the sensor device hardware obtained by formula (2) is as follows:

[0018] S i (λ)=t i (λ)q(λ) (3);

[0019] Where, t i (λ) is the spectral response filter, i = R, G, B, q(λ) is the quantum efficiency of the sensor material;

[0020] Under ideal conditions, it can be assumed that the spectral transmission rate p(λ) is 1. The reflectivity at this time is named ideal reflectivity, and the ideal reflectivity in the red, green, and blue (R, G, B) channels can be calculated using formula (4):

[0021]

[0022] Where r R 、r G and r B The reflectance is measured by the instrument under the red, green and blue (R, G, B) channels, r R,1 、r G,1 and r B,1 are the ideal reflectances under the red, green, and blue (R, G, B) channels respectively;

[0023] The theoretical first-order correction equation shown in formula (4) is obtained by fitting the reflectance of natural colored cotton and dyed cotton to the ideal reflectance, and the data before and after correction are respectively as follows: Figure 1 and Figure 2 As shown in the figure, the noise of optical components such as lens and the influence of surface iris are ignored. In actual situations, these factors will directly affect the measurement of color fiber whisker curve and the whisker curve of each color fiber in the mixed color fiber, that is, the spectral transmission rate p(λ) is not equal to 1. Therefore, a secondary correction should be made to the reflectivity. The influence of reflectivity on the accurate whisker curve calculation of the optical model is analyzed. The R ∞ The single color channel with a concentration less than 10% is used as the reference channel, and a new algorithm for the surface density distribution of the layered fiber assembly is derived to calculate the whisker curve of each channel:

[0024]

[0025] Where W r,ij is the relative surface density of the pixel at row i and column j in the whisker bundle transmission image; W ij is the optical density of the pixel at row i and column j in the whisker bundle transmission image; W max R is the optical density of the thickest part of the sample in the whisker transmission image; ∞ is the reflectivity of colored cotton fibers or other colored polymer objects when they are piled to infinite thickness, T ij is the transmittance of the pixel at row i and column j in the projected image measured and collected in step 1, T max is the transmittance at the thickest part of the sample in the projected image measured and collected in step 1;

[0026] Iteration is performed with the goal of minimizing the deviation of the whisker curve of each channel to obtain the estimated reflectance of other high-reflectivity channels. Based on these trends, the secondary correction equation corresponding to each channel can be fitted, and then the secondary correction reflectance under the red, green, and blue (R, G, B) channels can be calculated using formula (6):

[0027]

[0028] Where r R,1 、r G,1 and r B,1 are the ideal reflectances in the red, green, and blue (R, G, B) channels, r R,2 、r G,2 and r B,2 They are the secondary corrected reflectance under the red, green and blue (R, G, B) channels respectively;

[0029] Step 3: Obtaining beard clump curve and fluffiness correction:

[0030] After the random beards at both ends of colored cotton fibers are clamped by the clips, the fluffiness of the fibers on both sides will change, resulting in a deviation between the optical fiber amount and the actual fiber amount. In order to obtain a more accurate head-end signal of the beard curve, the relationship between the optical fiber amount and the fiber density must be studied. The relationship between the optical fiber amount and the fiber density under different fluffiness is studied. According to the different opening states of the combed dyed cotton strips and the lint bulk fibers of natural colored cotton, the relationship models between the optical fiber amount and the actual fiber amount in the beard curves of the combed dyed cotton strips and the lint bulk fibers of natural colored cotton are constructed as shown in formulas (7) and (8). The four combed dyed cotton strips of blue, yellow, red and green and the lint bulk fibers of the four natural colored cotton strips of white, light green, green and dark green are fitted, and the fitting curves are obtained as shown in formulas (7) and (8). Figure 3 、 4 As shown;

[0031] Formulas (7) and (8) are used to establish a bulkiness regression model to correct the bulkiness of the fiber beard:

[0032] P=-0.5024*x 3 +0.4853*x 2 +1.0142*x-0.0017 (7)

[0033] P1=-0.3696*x1 3 +0.0805*x1 2 +1.2909*x1-0.0006 (8)

[0034] Where x and x1 are the R channel optical relative fiber mass of dyed cotton and natural colored cotton in the random beard image after reflectivity correction, and P and P1 are the actual fiber mass of dyed cotton and natural colored cotton after bulk correction, respectively.

[0035] Preferably, in step 1, the infinite thickness, that is, the thickness is 0.5 to 3 cm.

[0036] Preferably, in step 1, colored cotton fibers with a mass of 0.2 to 0.6 g are used to make double-ended random beard bundles.

[0037] Preferably, in step 1, the three monochromatic lights of red, green and blue (R, G, B) can be replaced by monochromatic wave channels of different wavelengths.

[0038] Preferably, in step 1, the colored cotton fibers may be replaced by various fibers such as colored wool fibers, colored cashmere fibers, colored kapok fibers, colored chemical fibers or colored linen fibers, or colored polymer layered stacking materials.

[0039] Beneficial effects of the present invention:

[0040] The present invention provides a monochromatic light optical measurement method for colored cotton fiber whisker curves based on a secondary reflectivity estimation algorithm and fluffiness optical correction. The method first obtains fiber reflection signals and fiber whisker transmission signals using red, yellow, and blue three-channel monochromatic light. The reflectivity of colored cotton fibers with a fixed packing density is then corrected using a secondary monochromatic light reflectivity estimation optical algorithm. The optical linear density distribution curve of random double-ended whiskers is then calculated using the transmittance signal obtained from the whisker image, the corrected reflectivity signal, and an optical surface density algorithm. Finally, the fluffiness optical correction algorithm for combed cotton fibers or lint loose fibers is used to calculate the colored cotton fiber whisker curve. This method can be used to calculate the infinite thickness reflectivity of colored fiber whiskers and obtain accurate colored fiber mass distribution, whisker curve, and fiber length distribution, thereby solving the problem of ideal R caused by noise from optical components such as lenses and surface iris. ∞ The difficulties of measurement deviation and calculation deviation of optical fiber quantity distribution caused by changes in the fluffiness of fiber aggregates squeezed have been overcome. The problem that the American cotton fiber length measurement system HVI, which has a high usage rate among domestic measurement systems, cannot measure colored fibers has been overcome. By simply collecting the optical image information of the material, the infinite thickness reflectivity of the accurate mixed-color material can be calculated and the accurate whisker curve can be obtained, which is in good agreement with the reference curve measured by the international automated cotton fiber length standard test system AFIS.

[0041] The features and advantages of the present invention will be described in detail through embodiments with reference to the accompanying drawings.

Brief Description of the Drawings

[0042] Figure 1 This is the data of natural colored cotton and dyed cotton before correction;

[0043] Figure 2 The data of natural colored cotton and dyed cotton are after the first level correction;

[0044] Figure 3 is the fitting curve of dyed cotton fiber after carding;

[0045] Figure 4 is the fitting curve of the lint bulk fiber of natural colored cotton;

[0046] Figure 5 This is a physical picture of the carded and dyed cotton fiber used in Example 1;

[0047] FIG6 is a comparison of the beard clump curves of the carded dyed cotton fiber before and after correction in Example 1 and the beard clump curve obtained by the international standard method AFIS;

[0048] Figure 7 This is a Bland-Altman plot of the dyed sliver fiber after combing in Example 1;

[0049] Figure 8This is a physical picture of the natural colored cotton lint bulk fibers used in Example 2;

[0050] FIG9 is a comparison of the natural colored cotton lint bulk fiber beard clump curves before and after correction in Example 2 and the beard clump curves obtained using the international standard method AFIS;

[0051] Figure 10 This is the Bland-Altman plot of the natural colored cotton lint bulk fibers of Example 2. [Specific implementation method]

[0052] Example 1: Accuracy of the correction equation for the beard clump curve of dyed cotton sliver after combing:

[0053] Use 4 different colors of combed and dyed cotton strips (actual picture as shown Figure 5 As shown), they are numbered (a), (b), (c) and (d) and a verification experiment is carried out. Specifically, the above-mentioned dyed cotton strips are made into layered material samples when they are stacked to infinite thickness and the reflection images (pixels are 1000 dpi) are measured and collected under uniform red, green and blue monochromatic light illumination. Then, 0.6g of dyed cotton fiber is taken to make double-ended random beard clumps (the beard clumps are made using the method disclosed in the patent announcement number CN102645166B) and the transmission images (pixels are 1000 dpi) are measured and collected under uniform red, green and blue monochromatic light illumination. For the reflection image, the ideal reflectivity of the test sample is first calculated according to formula (4), and then the ideal reflectivity is corrected by a second-order approximation according to formula (6). For the transmission image, formula (5) is used to obtain the transmittance signal and the corrected reflectivity from the transmission image to calculate the beard clump curve, and then formula (7) is used to obtain the fluffiness correction beard clump curve.

[0054] The four groups of colored cotton fiber beard clump curves before and after correction (a), (b), (c) and (d) were compared with the beard clump curves obtained by the international standard method AFIS. The results are as follows: Figures 6a to 6d The results show that the deviation between the corrected whisker curve and the standard curve is smaller.

[0055] Then, the agreement between AFIS and random whisker bundle imaging fiber map was evaluated by Bland Altman. Figure 7 , most of the data points in the figure are distributed between the 95% confidence upper limit (the first line from top to bottom) and the lower limit (the third line from top to bottom), indicating that within the 95% confidence interval, the two methods are in good agreement and there is no significant difference. Figure 7 In the figure, the horizontal and vertical axes represent the mean and difference of the RBIM measurement value (M) and the AFIS reference value (R), respectively, and the second line from the top to the bottom represents the difference between the two average values.

[0056] The weight-average length is calculated based on the relationship between the whisker curve F(l) of random whiskers and the weight-length and compared with the AFIS results. The formula is as follows:

[0057] L w =2∫F(l)dl (9);

[0058] The results are shown in Table 1. It can be seen from the table that the deviation between the weight average length measured by the random whisker cluster imaging method and the measurement value of AFIS is very small, and they have high consistency.

[0059] Table 1 Comparison of weight average length of two test methods

[0060]

[0061] Example 2: Investigation of the accuracy of the correction equation on the beard clump curve of natural colored cotton lint bulk fibers:

[0062] Using 4 different colors of natural colored cotton lint loose fibers (actual picture as shown Figure 8 As shown), they are numbered (a), (b), (c) and (d) and a verification experiment was carried out. Specifically, the above-mentioned natural colored cotton lint bulk fibers were made into layered material samples when they were piled to infinite thickness and the reflection images (pixels were 1000 dpi) were measured and collected under uniform red, green and blue monochromatic light illumination. Then, 0.6g of natural colored cotton lint bulk fibers were taken to make double-ended random whiskers (the whiskers were made using the method disclosed in the patent announcement number CN102645166B) and the transmission images (pixels were 1000 dpi) were measured and collected under uniform red, green and blue monochromatic light illumination. For the reflection image, the ideal reflectivity of the test sample was first calculated according to formula (4), and then the ideal reflectivity was corrected by a second-order approximation according to formula (6). For the transmission image, formula (5) was used to obtain the transmittance signal and the corrected reflectivity from the transmission image to calculate the whisker curve, and then formula (8) was used to obtain the fluffiness correction whisker curve.

[0063] The four groups of colored cotton fiber beard clump curves before and after correction (a), (b), (c) and (d) were compared with the beard clump curves obtained by the international standard method AFIS. The results are as follows: Figures 9a to 9d The results show that the deviation between the corrected whisker curve and the standard curve is smaller.

[0064] Then, the agreement between AFIS and random whisker bundle imaging fiber map was evaluated by Bland Altman. Figure 10 , most of the data points in the figure are distributed between the 95% confidence upper limit (the first line from top to bottom) and the lower limit (the third line from top to bottom), indicating that within the 95% confidence interval, the two methods are in good agreement and there is no significant difference. Figure 10 In the figure, the horizontal and vertical axes represent the mean and difference of the RBIM measurement value (M) and the AFIS reference value (R), respectively, and the second line from the top to the bottom represents the difference between the two average values.

[0065] The weight-average length was calculated according to formula (9) and compared with the AFIS results. The results are shown in Table 2. It can be seen from the table that the measurement parameters of the double-ended whisker image method are highly consistent with the measurement parameters of the AFIS.

[0066] Table 2 Comparison of weight average length of two test methods

[0067]

[0068]

[0069] The above embodiments are intended to illustrate the present invention, not to limit the present invention. Any solution that is a simple transformation of the present invention falls within the protection scope of the present invention.

Claims

1. An optical measurement method for colored cotton fiber beard curve based on approximate secondary reflectivity and bulk correction, characterized in that: The steps include: Step 1: Image measurement and acquisition: Obtain a layered material sample made of colored cotton fibers or other colored polymer objects stacked to infinite thickness and measure and collect the reflection image under uniform red, green, and blue monochromatic light illumination. Then, use the colored cotton fibers to make a double-ended random whisker bundle and measure and collect the transmission image under uniform red, green, and blue monochromatic light illumination. Step 2: Correction of optical algorithm for secondary estimation of monochromatic light reflectance: The instrument measured reflectance in the red, green and blue channels is calculated using formula (1): r i ＝U i / 255 (1)； Where U i are the values ​​of the R, G, and B channels obtained from the reflection image measured and collected in step 1, i = R, G, B; The ideal reflectivity in red, green and blue channels is calculated using formula (2): Where r R 、r G and r B The reflectance is measured by the instrument under the red, green and blue channels, r R,1 、r G,1 and r B,1 are the ideal reflectances in red, green and blue channels respectively; The secondary correction reflectance in the red, green and blue channels is calculated using formula (3): Where r R,1 、r G,1 and r B,1 are the ideal reflectivity in red, green and blue channels, r R,2 、r G,2 and r B,2 They are the secondary corrected reflectance in red, green and blue channels respectively; Step 3: Obtaining beard clump curve and fluffiness correction: Formulas (7) and (8) are used to establish a bulkiness regression model to correct the bulkiness of the fiber beard: P=-0.5024*x 3 +0.4853*x 2 +1.0142*x-0.0017 (7) P1=-0.3696*x1 3 +0.0805*x1 2 +1.2909*x1-0.0006 (8) Where x and x1 are the R channel optical relative fiber mass of dyed cotton and natural colored cotton in the random beard image after reflectivity correction, and P and P1 are the actual fiber mass of dyed cotton and natural colored cotton after bulk correction, respectively.

2. The optical measurement method for colored cotton fiber beard tuft curve based on approximate secondary reflectivity and bulk correction according to claim 1, characterized in that: In step 1, the infinite thickness is 0.5 to 3 cm.

3. The optical measurement method for colored cotton fiber beard tuft curve based on approximate secondary reflectivity and bulk correction according to claim 1, characterized in that: In the step 1, colored cotton fibers with a mass of 0.2 to 0.6 g are used to make double-ended random beard bundles.

4. The optical measurement method for colored cotton fiber beard tuft curve based on approximate secondary reflectivity and bulk correction according to claim 1, characterized in that: In step 1, the three monochromatic lights of red, green and blue can be replaced by monochromatic wave channels of different wavelengths.

5. The optical measurement method of colored cotton fiber beard tuft curve based on approximate secondary reflectivity and bulk correction according to claim 1, characterized in that: In step 1, the colored cotton fibers may also be replaced by various fibers such as colored wool fibers, colored cashmere fibers, colored kapok fibers, colored chemical fibers or colored linen fibers, or colored polymer layered stacking materials.

Images