CN112268905B - Image detection method of short fiber content in textile fibers - Google Patents

Abstract

The invention proposes an image detection method for short fiber content in textile fibers, which includes the following steps: Step S1: prepare a double-ended whisker sample from the sample, and scan to obtain its light-transmitting image; The light and linear density conversion method obtains the unilateral whisker curve; step S3: calculates and obtains a detection value of the short fiber content SFC _w (α) of the sample for the unilateral whisker curve; step S4: repeats steps S1-step S3, for the same A plurality of double-ended whisker samples were prepared from the sample, each SFCw(α) detection value was calculated separately, and the average value was calculated as the final detection value of the short fiber content of the sample. There is no theoretical systematic error between the calculated value of the short fiber content calculated by the present invention and the true value; the algorithm principle of the present invention is clear, the calculation amount is small, and it is convenient and fast; the scheme of the present invention can directly use mature commercialized general-purpose equipment to detect the short fiber content , Compared with other special automation instruments, it greatly reduces the equipment purchase cost, and the use and maintenance costs are extremely low.

Description

Image detection method for short fiber content in textile fibers

Technical Field

The present invention belongs to the field of textiles, and in particular relates to an image detection method for the short fiber content in textile fibers, or a rapid and low-cost detection method for the short fiber content in textile fiber raw materials or spinning semi-finished products, which is suitable for the detection of the short fiber content of cotton, chemical fiber staple fibers, plush fibers and other textile natural fibers.

Background Art

Textile fibers often have different lengths. The length of the same batch of fiber raw materials is often characterized by length distribution and several length indicators. Among them, the content of fibers with a length less than a certain limit is called short fiber content, which is an important quality indicator of textile fibers and directly affects the tensile properties of yarns, surface hairiness and fabric style, etc. Therefore, it is an important basis for process design, equipment selection, product grading and trade pricing.

Manual testing methods for short fiber content include hand-arranged plot test method (Baye plot method), roller fiber length test method, comb fiber length test method, etc. Manual testing methods generally have problems such as complex sample preparation, slow testing speed, and low stability. In addition, they require high skill proficiency of testers, and the data differences between different testers are large. It is difficult to meet the needs of standardization and rapidity of the modern textile industry.

At present, common automated testing instruments include the camera method (represented by HVI instrument), the single fiber rapid test method (represented by AFIS instrument), the capacitive fiber length test method (represented by Almeter100 instrument) and the line-by-line scanning image method (represented by OFDA4000 instrument). When using HVI for length measurement, a certain amount of loose fibers are first placed in a cylindrical sampler with holes around the periphery, and then a linear clamp is rotated along the outer wall of the sampler to hook and clamp the fibers exposed from the holes, combed by a brush to form a measurable beard clump, and sent to the photoelectric detection area; the fiber fragments near the beard clamping line are excluded from the detection area due to unclear entanglement, but short fibers are mainly present in this position, so HVI cannot directly measure the short fiber content, and can only calculate a short fiber characterization parameter - short fiber index (SFI) based on the measured fiber strength, maturity and other indicators combined with empirical equations. The AFIS system calculates the length distribution index of the fiber sample by measuring the length of thousands of single fibers. Its built-in opening components may cause fiber breakage when separating fibers, resulting in deviations in the measurement results. Almeter100 first uses a special manipulator to make a whisker clump with one end flat, then passes the whisker clump through the capacitive sensor at a constant speed, and calculates the change information of the number of fibers in each cross section from the root to the tip of the whisker clump based on the capacitance deviation caused by the change of the fiber amount in each cross section of the whisker clump, and finally makes a length distribution diagram and calculates various length indicators. OFDA4000 also needs to make a whisker clump with one end flat, and then based on CCD microscopic imaging technology, measure the number of fibers in the whisker clump every 5 mm cross section, and perform simulation interpolation processing at a spacing of 1 mm. Each sample tests 5 whisker clumps, and calculates the fiber length distribution in the whisker clump and its corresponding short fiber content and other length indicators based on the change of the number of fibers in each cross section from the root to the tip of the whisker clump.

The above-mentioned various automated measurement systems have the advantages of being fast, efficient, and labor-saving. However, each measurement system is very large in size, has complex operating mechanical and electrical parts, is expensive to manufacture, and requires a constant temperature and humidity chamber, which results in high maintenance costs.

In 2012, Wang Fumei and Wu Hongyan proposed a new method for fast and low-cost fiber length measurement (patent number: ZL201210106711.8), which is as follows: first, the bulk fiber sample to be tested is made into a fiber strip with straight, parallel and randomly arranged fibers, and then a special double-ended whisker sample is made, and then the transmission grayscale image of the whisker is obtained by using an optical imaging device, and the relationship curve between the fiber quantity and position on any cross section of the whisker is calculated from the image (referred to as the whisker curve), and finally the whisker curve is used to calculate four indicators such as weighted average length, main body length, quality length and length variation coefficient. This method can use low-cost and portable imaging equipment to accurately measure several length distribution indicators in spinning raw materials or fiber whiskers in a short time, which is largely due to its innovative double-ended whisker sample. When preparing the sample, use a clamp to randomly clamp any cross section of the fiber strip, ensure that the clamping line is perpendicular to the axis of the fiber strip, comb out the floating fibers that are not clamped, and obtain a double-ended whisker with two tapered ends. Compared with the other sample forms mentioned above, the double-ended whisker sample is not only quick to prepare, but also the fibers in the whisker are parallel and straight, with no undetectable blind spots. Therefore, all length information (including short fiber information that is most easily lost by other methods) can be reflected in the image, which makes it possible to accurately measure the short fiber content.

In 2013, Wang Fumei, Jin Jingye and Chen Fei proposed a method for calculating the short fiber content based on a double-ended whisker sample (patent number: ZL201310325921.0). First, some geometric characteristic values that may be related to the short fiber content are selected from the whisker curve, and then the short fiber content is measured by other methods, so as to establish a statistical regression equation or an empirical prediction equation based on batch tests. The prediction accuracy of this method depends on the type and quantity of samples in the previous batch test, and does not involve other fibers except cotton and kapok. The theoretical basis is weak, so that the prediction equation for the content of fibers below 16 mm does not contain 16 mm related variables but only 12.7 mm related variables, which is an illogical phenomenon. In 2016, Wang Fumei, Xu Bugao and Jin Jingye proposed a method for measuring and calculating the short fiber content based on a double-ended whisker step-by-step separation model (patent number: ZL201610033913.2). By gradually decomposing the whisker, the short fibers below a specific length are gradually separated, and their content is calculated. This method requires a large amount of calculation, and according to the theory of the stepwise separation model, the amount of short fibers separated is only gradually approaching the true value, and theoretically can never be equal to the true value; in addition, the whisker optical algorithm used is defective, resulting in the calculated whisker curve being in a "bald" shape, rather than the theoretical "pointed" shape, which will lead to the risk of systematic deviation in the test results. Afterwards, Wang Fumei, Chen Lijun, Wu Meiqin, and Zhao Lin proposed a new method for obtaining the linear density coefficient curve and the standard whisker curve (patent number: ZL201611230426.1), which helps to obtain a bilateral whisker curve that is closer to the theoretical shape, laying the foundation for the accurate extraction of various length indicators.

Summary of the invention

In view of the deficiencies and gaps in the prior art, the present invention provides an image detection method for the short fiber content in textile fibers, or a rapid and low-cost detection method for the short fiber content in textile fiber raw materials or spinning semi-finished products, which is suitable for the short fiber content detection of cotton, chemical fiber staple fibers, plush fibers and other textile natural fibers.

The present invention specifically adopts the following technical solutions:

An image detection method for short fiber content in textile fibers, characterized in that it comprises the following steps:

Step S1: preparing a double-ended whisker sample from a sample, and scanning to obtain a light transmission image thereof;

Step S2: Use the whisker transmittance and line density conversion method to extract the double-sided whisker curve F(L) from the transmittance image, and fold it along the vertical axis, and average the vertical coordinate values of the two points with symmetrical horizontal coordinate values on the curve to obtain the single-sided whisker curve F(L) to reduce the random error caused by the unevenness of the fiber itself; wherein the whisker transmittance and line density conversion method used is the method provided in "Chinese Patent ZL201611230426.1 Linear Density Coefficient Curve and Standard Whisker Curve Acquisition Method".

Step S3: applying formula (1) to the single-sided whisker curve F(L) to calculate a detection value of the short fiber content SFC _w (α) of the sample;

SFC _w (α)=m×[αF′(α)-F(α)+1]-n (1)

Wherein, α is the length limit of short fibers, and different countries and regions have different α values for different fibers; F(α) is the ordinate value of the single-end whisker curve at the abscissa value of α, F’(α) is the slope of the single-end whisker curve at the abscissa value of α, m and n are constants determined by the fiber type and fiber straightness, and m = 0 to 4, n = 0 to 9;

Step S4: Repeat steps S1 to S3 to prepare multiple double-ended whisker tuft specimens for the same sample, calculate and obtain each SFCw(α) test value respectively, and obtain the average value as the final test value of the short fiber content of the sample.

Preferably, the sample is cotton or chemical fiber staple fiber or plush fiber.

Preferably, in step S3, the value of α is 8 mm to 30 mm.

The present invention and its preferred embodiment have the following beneficial effects: (1) An overall scheme is constructed based on the whisker clump transmittance and line density conversion algorithm, so that the whisker clump curve extracted from the image is more ideal and closer to the theoretical whisker clump curve shape, ensuring that the calculation result of the short fiber content algorithm of the present invention is accurate and effective; (2) There is no theoretical systematic error between the calculated short fiber content value obtained by the present invention and the true value; (3) The algorithm principle of the present invention is clear, the calculation amount is small, and it is convenient and fast; (4) The scheme of the present invention can directly use mature commercial general equipment to detect the short fiber content. Compared with other special automated instruments, it greatly reduces the equipment procurement cost, and the use and maintenance costs are extremely low.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

FIG1 is a schematic diagram of a double-ended beard tuft sample (cotton fiber) provided in an embodiment of the present invention;

FIG2 is a schematic diagram of a light transmission scanning image of a double-ended whisker bundle according to an embodiment of the present invention;

FIG3 is a schematic diagram of converting a double-sided whisker cluster curve into a single-sided whisker cluster curve according to an embodiment of the present invention;

FIG4 is a schematic diagram of the calculation principle of formula (1) according to an embodiment of the present invention.

DETAILED DESCRIPTION

In order to make the features and advantages of this patent more obvious and easy to understand, the following embodiments are specifically described in detail as follows:

This embodiment provides an image detection method for the short fiber content in textile fiber raw materials and spinning semi-finished products, the steps of which are as follows:

The first step is to prepare a double-ended whisker sample: a certain amount of fiber is taken from a laboratory sample (fiber raw material or semi-finished sliver of spinning process) to make a whisker with straight fibers and uniform distribution, and a special clamp is used to vertically clamp any cross section of the whisker, and the fibers not held on both sides of the clamping line are combed out to obtain a double-ended whisker, as shown in FIG1 ;

In the second step, the double-ended whisker tuft is placed in a transmission scanner and scanned to obtain its image, as shown in Figure 2; then, the whisker tuft optics and line density conversion algorithm are used to extract the double-sided whisker tuft curve F(L), which is a curve showing the relationship between the relative fiber amount (number of fibers) and the position on any cross section of the whisker tuft, and L is the position of any cross section of the whisker tuft relative to the clamping line (L=0); then, the double-sided whisker tuft curve is folded in half along the ordinate axis, and the ordinate values of the two points on the curve whose abscissa values are symmetrical are averaged to obtain the single-sided whisker tuft curve F(L) to reduce the random error caused by the unevenness of the fibers themselves, as shown in Figure 3; in this embodiment, the whisker tuft transmittance and line density conversion method used is the method provided in "Chinese Patent ZL201611230426.1 Linear Density Coefficient Curve and Standard Whisker Tuft Curve Acquisition Method".

The third step is to use formula (1) to calculate a detection value of the short fiber content SFCw(α) of the laboratory sample, where α is the length limit of the short fiber, generally 8 mm to 30 mm; F(α) is the ordinate value of the single-end whisker curve at the abscissa value of α, F’(α) is the slope of the single-end whisker curve at the abscissa value of α, m and n are constants determined by the fiber type and fiber straightness, and m = 0 to 4, n = 0 to 9.

SFC _w (α)=m×[αF′(α)-F(α)+1]-n (1)

Step 4: Repeat the above three steps to make several double-ended whisker bundles for the same laboratory sample, calculate a SFCw(α) test value for each of them, and calculate the average of these test values, which is the final test value of the short fiber content of the laboratory sample.

As shown in FIG4 , the technical principle of this embodiment is as follows:

According to the published research literature, the relationship between the whisker curve F(L) and the fiber length weight frequency density function _pw (l) of the laboratory sample is:

p _w (l) = lF″(l) (2)

Where l is the fiber length. According to the definition of short fiber content, the weight proportion of short fibers in laboratory samples SFCw(α) is calculated as:

Where α is the short fiber length limit. Combining formulas (2) and (3), and applying the method of integration by parts, we can obtain:

Theoretically, the short fiber content (weight ratio) of the laboratory sample can be calculated by applying formula (4) to the whisker curve, but the premise is that the fibers in the whisker are completely straight. In fact, natural textile fibers inevitably have a certain degree of twist or curl, and different types of fibers have different degrees of straightness, which leads to the difference between the short fiber content calculated by formula (4) and the benchmark value measured by other existing methods. In order to correct the deviation, this embodiment establishes a mathematical model, conducts large-scale tests on each fiber type, compares the calculation results of formula (4) with the benchmark values measured by other methods, and uses statistical methods to determine the buckling correction coefficients m and n of various fibers to obtain the final calculation formula (1) for the short fiber content (weight ratio).

The following is further explained with reference to specific examples.

Example 1

According to Chinese standards, the short fiber length limit of fine-staple cotton is α = 16 mm. Taking the fiber buckling coefficient of fine-staple cotton as m = 0.93 and n = 1.66, formula (1) is converted to

SFC _w (16)＝0.93×[16F′(16)-F(16)+1]-1.66 (5)

Three kinds of fine cotton samples with different short fiber contents were randomly selected, which were from the United States, Mexico and India. Four double-ended beard tuft specimens were made for each sample. Formula (5) was applied to each specimen to obtain a test value. The average of the test values of the four specimens was taken as the final test value and compared with the short fiber content measured by the AFIS instrument, as shown in Table 1. Obviously, the present invention has a high consistency with the benchmark method.

Table 1 Comparison of the short fiber content of fine cotton detected by the present invention and the detection value of the AFIS benchmark method

Example 2

In recent years, kapok fiber, as an emerging natural fiber raw material with antibacterial and mite-repellent properties and light weight and warmth retention, has been increasingly used in textiles. The average length of kapok fiber is relatively short, so referring to the commonly used cotton short fiber limit internationally, α is set to 12.7 mm; in addition, kapok has good straightness and basically no bending or curling. Taking the fiber bending coefficient m = 1.46 and n = 6.37, formula (1) is converted to

SFC _w (12.7)＝1.46×[12.7F′(12.7)-F(12.7)+1]-6.37 (6)

Three kapok fiber samples with different short fiber contents were randomly selected, which were from Indonesia and Hainan and Panzhihua in my country. Four double-ended beard tuft samples were made from each sample. Formula (6) was applied to each sample to obtain a test value. The average of the test values of the four samples was taken as the final test value and compared with the short fiber content measured according to GB/T16257-2008 "Test for short fiber length and length distribution of textile fibers: Single fiber measurement method". As shown in Table 2, it is shown that the present invention also has a high consistency with the benchmark method in measuring the short fiber content of new fibers such as kapok fibers.

Table 2 Comparison of kapok short fiber content detected by the present invention and the measured value by the benchmark method

Example 3

Wool fiber is long and has more curls. The short fiber length limit generally used in my country is α=30 mm. In the present invention, the fiber bending coefficient m=0.62 and n=2.29 are taken, and formula (1) is converted to

SFC _w (30)＝0.62×[30F′(30)-F(30)+1]-2.29 (7)

Three wool fiber laboratory samples with different short fiber contents were randomly selected, which were from Inner Mongolia, Australia and New Zealand, and four whisker clump samples were made for each sample. After the whisker clump curve was extracted, the test value was obtained by applying formula (7). The average of the test values of the four samples was taken as the final test value and compared with the short fiber content (weight ratio) measured by the Almeter instrument, as shown in Table 3, which shows the accuracy of the present invention in measuring the short fiber content of wool.

Table 3 Comparison of wool short fiber content detected by the present invention and the measured value by the benchmark method

sample The short-hair rate measured by the present invention/% Short fiber rate measured by Almeter/% Difference rate/% Inner Mongolia wool 4.5 4.1 8.89% Australian Wool 2.8 3.0 -7.14% New Zealand Wool 4.0 3.7 7.50%

This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive other forms of short fiber content image detection methods under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of the present invention should be covered by this patent.

Claims (3)

1. an image detection method of short fiber content in textile fibers, is characterized in that, comprises the following steps: Step S1: Prepare a double-ended whisker sample from the sample, and scan to obtain its light-transmitting image; Step S2: Extract the double-sided whisker curve F(L) from the light-transmitting image by using the whisker transmission and linear density conversion method, fold the bilateral whisker curve in half along the ordinate axis, and pair the two sides with symmetrical abscissa values on the curve. The ordinate values of the points are averaged to obtain the unilateral whisker cluster curve F (L); the unilateral whisker cluster curve F (L) is the relational curve between the relative fiber amount and position on any cross-section of the whisker cluster, and L is The position of any cross-section of the whiskers relative to the clamping line; Step S3: apply the formula (1) to the unilateral whisker curve F(L) to calculate and obtain a detection value of the sample short fiber content SFC _w (α); SFC _w (α)=m×[αF′(α)-F(α)+1]-n(1) Among them, α is the length limit of the short fiber; F(α) is the ordinate value of the single-ended whisker curve at the abscissa value α, and F'(α) is the single-ended whisker curve at the abscissa value α The slope of , m, n are constants determined by fiber type and fiber straightness, and m=0~4, n=0~9; Step S4: Repeat steps S1-S3, make multiple double-ended hair tuft samples for the same sample, calculate and obtain each SFCw(α) detection value, and calculate the average value as the final detection value of the short fiber content of the sample. 2. The image detection method for short fiber content in textile fibers according to claim 1, characterized in that: the sample is cotton or chemical staple fiber or plush fiber. 3. The image detection method of short fiber content in textile fibers according to claim 1, characterized in that: in step S3, the value of α is 8 mm to 30 mm.

Images