Yarn residue detection method, device and storage medium of yarn management machine

Abstract

The present invention discloses a yarn residue detection method of a yarn management tube machine, comprising the following steps: step 1, collecting the RGB picture of the current yarn tube; step 2, pre-processing the picture, step 2.1, Gaussian filtering the picture; step 2.2, performing RGB channel separation; step 2.3, Gaussian filtering the separated RGB channel picture; step 2.4, performing canny edge detection on the filtered picture again; step 3, performing residual yarn detection on the yarn tube, step 3.1, judging whether the front half of the yarn tube contains yarn; step 3.2, extracting the color of the yarn tube; step 3.3, judging whether the back half of the yarn tube contains yarn. In the above technical scheme, the influence of factors such as the intensity of the ambient light source, the cleanliness of the yarn tube, the placement position of the yarn tube, the saturation of the yarn, the density of the yarn and the like on the residual yarn detection can be greatly reduced, thereby achieving the advantages of high efficiency and high detection accuracy.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a yarn residue detection method, a device and a storage medium thereof for a yarn management machine. Background Art

[0002] In the textile industry, traditional yarn residue detection is done through manual detection and classification, which increases production costs and reduces production efficiency. In addition, the error rate of manual identification is high, especially when it is close to the physiological limit of the human eye, which greatly reduces the development of the traditional textile industry.

[0003] In recent years, with the development of the field of image processing, image processing has also been used in yarn residue detection. The detection of yarn residue through image processing has a high degree of automation, reduced labor costs, greatly improved production efficiency, and no physiological limit. However, the existing yarn tube image processing has great limitations in application, and has very high requirements for the working environment. It is easily affected by the intensity of the ambient light source, the cleanliness of the yarn tube, the placement of the yarn tube, the saturation of the yarn, the density of the yarn, etc., which can easily cause misjudgment and omission of the yarn residue detection of the yarn tube, which greatly limits the application of image processing in yarn residue detection. Summary of the invention

[0004] In view of the deficiencies in the prior art, one of the purposes of the present invention is to propose a yarn residue detection method for a yarn management tube machine, which can greatly reduce the influence of factors such as the intensity of the ambient light source, the cleanliness of the yarn tube, the placement of the yarn tube, the saturation of the yarn, and the density of the yarn on the residual yarn detection, thereby achieving the advantages of high efficiency and high detection accuracy.

[0005] The second purpose is to propose a yarn residue detection device for a yarn management tube machine, which can greatly reduce the influence of factors such as the intensity of ambient light source, the cleanliness of the yarn tube, the placement of the yarn tube, the saturation of the yarn, and the density of the yarn on the residual yarn detection, thereby achieving the advantages of high efficiency and high detection accuracy.

[0006] In view of the deficiencies in the prior art, the present invention proposes a yarn residue detection device for a yarn tube management machine, which can greatly reduce the influence of factors such as the intensity of the ambient light source, the cleanliness of the yarn tube, the placement of the yarn tube, the saturation of the yarn, and the density of the yarn on the residual yarn detection, thereby achieving the advantages of high efficiency and high detection accuracy.

[0007] In order to solve the above technical problems, the technical solution of the present invention is:

[0008] A yarn residue detection method for a yarn management machine comprises the following steps:

[0009] Step 1: Collect the RGB image of the current bobbin

[0010] The interception of the RGB image is recorded as f, f(left, top, right, bottom) represents the interception of the collected f, left, top, right, bottom represent the values ​​of the left, top, right, and bottom of f respectively, let g(x, y) = f(left, top, right, bottom) represent the intercepted bobbin image, wherein x and y represent the ordinate and abscissa of the image respectively;

[0011] Step 2: Preprocess the image

[0012] Step 2.1, perform Gaussian filtering on the image;

[0013] Step 2.2, separate the RGB channels;

[0014] Step 2.3, perform Gaussian filtering on the separated RGB channel image;

[0015] Step 2.4, perform canny edge detection on the filtered image;

[0016] Step 3: Detect the residual yarn on the bobbin

[0017] Step 3.1, determine whether the front half of the bobbin contains yarn;

[0018] Step 3.2, extracting the color of the yarn tube;

[0019] Step 3.3, determine whether the rear half of the bobbin contains yarn.

[0020] Preferably, in step 2, the intercepted bobbin image g(x, y) is preprocessed.

[0021] In step 2.1, Gaussian filtering is performed on g(x, y) according to the following formula:

[0022] g σ (x, y) = g(x, y)*G σ

[0023] Where * represents the convolution operation, G σ is a two-dimensional Gaussian kernel with standard deviation σ, defined as:

[0024]

[0025] In step 2.2, g is calculated according to the following formula: σ (x, y) performs RGB channel separation:

[0026] g b =g σ [0]

[0027] g g =g σ [1]

[0028] g r =g σ [2]

[0029] where g b Indicates the B channel of the bobbin image, g g Represents the G channel of the bobbin image, g r The R channel represents the bobbin image;

[0030] In step 2.3, repeat step 2.1 to perform Gaussian filtering on the separated RGB channel images respectively:

[0031] g bσ =g b *G σ

[0032] g gσ =g g *G σ

[0033] g rσ =g r *G σ .

[0034] Preferably, the canny edge detection is used to detect the image g σ (x, y) for edge detection, the specific method is as follows:

[0035] First, calculate the gradient value and gradient direction of the image, and calculate the gradient value G(x, y) and gradient direction θ by the following formula:

[0036]

[0037]

[0038] where g x (x, y) 2 , g y (x, y) 2 Respectively represent g σ (x, y) returns the derivative values ​​in the horizontal and vertical directions through Sobel;

[0039] The gradient value G(x, y) is non-maximum suppressed according to the following formula:

[0040]

[0041] Where T1 represents the threshold of non-maximum suppression, filtering out noise;

[0042] According to the following formula, G T1 (x, y) performs double threshold screening:

[0043]

[0044] Where maxVal represents the high threshold, minVal represents the low threshold, and edge represents the pixel of the image to be retained, otherwise it is cleared. canny (x, y) represents the edge image after filtering.

[0045] Preferably, the method for determining whether the front half of the bobbin contains yarn is as follows:

[0046] According to the following formula, the edge image G canny (x, y) performs gradient statistics and performs threshold judgment to determine whether the first half contains yarn:

[0047]

[0048] Among them G canny .width indicates edge image G canny The width of (x, y),

[0049] If there exists j1 such that If T2 is the threshold of whether the first half contains yarn, then the first half of the bobbin contains yarn, otherwise the first half of the bobbin does not contain yarn, where j1∈(0, len(h(y))-α) represents the starting coordinates of the sliding summation window, α1 represents the length of the sliding summation window, and len(h(y)) represents the length of h(y).

[0050] Preferably, the method for extracting the color of the yarn tube is as follows:

[0051] For the filtered bobbin image g σ (x, y) is grayed out according to the following formula:

[0052] Gray=R*0.299+G*0.587+B*0.114

[0053]

[0054] Gray represents the grayscale formula. σ Each pixel in (x, y) is grayed out according to the formula to obtain g Gray (x, y);

[0055] Identify the reflection and dirty spots of the yarn tube;

[0056] Extract the tube body color.

[0057] Preferably, the method for judging the reflection and dirty spots of the yarn tube is as follows:

[0058] Use the following formula to determine whether there are reflective points and extract the grayscale values ​​of reflective points and dirty points:

[0059] Ref=MAX(g Gray (x, y)

[0060] Dir=MIN(g Gray (x, y)

[0061]

[0062] Where Ref represents the gray value of the reflective point, Dir represents the gray value of the dirty point, and T3 represents the threshold value of whether there is a reflective point;

[0063] If the grayscale value of the reflective point and the grayscale value of the dirty point are greater than the threshold T3, it means there is a reflective point, otherwise there is no reflective point.

[0064] Preferably, the method for extracting the tube body color is as follows:

[0065] After separating the RGB channels, g bσ , g gσ , g rσ Each pixel point (x∈(0, g σ .width), y∈(0, g σ .height / 2),) color extraction is performed according to the following formula:

[0066] If there are reflective spots:

[0067]

[0068] If there is no reflective point

[0069]

[0070] The R component of the tube body color is MAX(R_NUM), where R_NUM represents the number of pixels of the R color component between 0 and 255. σ .width indicates the width of the image, g σ .height indicates the length of the image, T4 indicates the threshold of whether a pixel is valid, T5 indicates the threshold of whether a pixel is a reflective point, and the B and G components are operated according to the above operation. Similarly, the RGB color value of the tube body can be obtained.

[0071] Preferably, the method for judging whether the second half of the bobbin contains yarn is as follows:

[0072] Match the pixel points in each column of the second half of the tube with the RGB value of the tube body and accumulate them:

[0073]

[0074] Where T6 represents the threshold value of whether the pixel point successfully matches the tube body RGB, and Mat(y) represents the number of pixels in each column that match the tube body;

[0075] Mat(y) is accumulated using a sliding window according to the following formula:

[0076]

[0077] Where j2∈(0,len(Mat)-α2) represents the starting coordinates of the sliding window, α2 represents the length of the sliding window, T7 represents the threshold of the sliding window to determine whether there is yarn, len(Mat) represents the length of the matching array Mat(y), slid(j2) represents the accumulated value of Mat(y) after passing through the sliding window, if the accumulated value is greater than the threshold T7, it means there may be yarn,

[0078] The residual yarn detection of the yarn-containing tube is carried out by the following formula:

[0079]

[0080] Where len(slid) represents the length of slid, and T8 represents the threshold for judging whether there is yarn;

[0081] If the number of times the possible yarn appears is greater than the threshold value T8, it is determined that the bobbin contains yarn, otherwise it does not contain yarn, and the determination of whether the bobbin contains yarn is completed.

[0082] The present invention also discloses a yarn residue detection device for a yarn management tube machine, comprising a memory, a processor and a program stored in the memory, wherein the program is configured to be executed by the processor, and when the processor executes the program, the yarn residue detection method for the yarn management tube machine is implemented.

[0083] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the yarn residue detection method of the yarn management machine is realized.

[0084] The present invention has the following characteristics and beneficial effects:

[0085] By adopting the above technical solution and using the image processing method to quickly and accurately detect the residual yarn on the bobbin, the influence of factors such as the intensity of the ambient light source, the cleanliness of the bobbin, the placement of the bobbin, the saturation of the yarn, and the density of the yarn on the residual yarn detection can be reduced. It has the advantages of high efficiency and high detection accuracy. DETAILED DESCRIPTION

[0086] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0087] Example 1

[0088] This embodiment provides a yarn residue detection method of a yarn management machine, comprising the following steps:

[0089] Step 1: Collect the RGB image of the current bobbin

[0090] The interception of the RGB image is recorded as f, f(left, top, right, bottom) represents the interception of the collected f, left, top, right, bottom represent the values ​​of the left, top, right, and bottom of f respectively, let g(x, y) = f(left, top, right, bottom) represent the intercepted bobbin image, where x and y represent the ordinate and abscissa of the image respectively;

[0091] Step 2: Preprocess the image

[0092] Step 2.1, perform Gaussian filtering on the image;

[0093] Step 2.2, separate the RGB channels;

[0094] Step 2.3, perform Gaussian filtering on the separated RGB channel image;

[0095] Step 2.4, perform canny edge detection on the filtered image;

[0096] Step 3: Detect the residual yarn on the bobbin

[0097] Step 3.1, determine whether the front half of the bobbin contains yarn;

[0098] Step 3.2, extracting the color of the yarn tube;

[0099] Step 3.3, determine whether the rear half of the bobbin contains yarn.

[0100] In the above technical solution, by collecting RGB images and capturing the values ​​of the left, right, top and bottom, the bobbin body domain is positioned, so that the placement of the bobbin and other factors have an impact on the residual yarn detection, thereby achieving the advantages of high efficiency and high detection accuracy. In addition, the amount of calculation is reduced, and the detection efficiency is improved while ensuring the detection accuracy.

[0101] Furthermore, in step 2, the intercepted bobbin image g(x, y) is preprocessed.

[0102] In step 2.1, Gaussian filtering is performed on g(x, y) according to the following formula:

[0103] g σ (x, y) = g(x, y)*G σ

[0104] Where * represents the convolution operation, G σ is a two-dimensional Gaussian kernel with standard deviation σ, defined as:

[0105]

[0106] In step 2.2, g is calculated according to the following formula: σ (x, y) performs RGB channel separation:

[0107] g b =g σ [0]

[0108] g g =g σ [1]

[0109] g r =g σ [2]

[0110] where g b Indicates the B channel of the bobbin image, g g Represents the G channel of the bobbin image, g r The R channel represents the bobbin image;

[0111] In step 2.3, repeat step 2.1 to perform Gaussian filtering on the separated RGB channel images respectively:

[0112] g bσ =g b *G σ

[0113] g gσ =g g *G σ

[0114] g rσ =g r *G σ .

[0115] In the above technical solution, the intercepted bobbin image g(x, y) is subjected to Gaussian filtering, smoothing filtering, noise removal, and weighted averaging of the entire image. The value of each pixel is obtained by weighted averaging the value of itself and other pixels in the neighborhood, thereby obtaining a better image edge, which lays the foundation for the subsequent detection accuracy.

[0116] In addition, after completing the Gaussian filter, the image is separated by the RGB channel, and the Gaussian filter is performed again after the separation is completed. It can be understood that the filter is performed again after separation, so that the noise can be better filtered out in detail, and a better image edge can be further obtained to ensure the accuracy of the later detection.

[0117] It can be understood that through separation and two Gaussian filtering, the influence of factors such as the intensity of the ambient light source, the cleanliness of the yarn tube, the saturation of the yarn, and the density of the yarn on the residual yarn detection can be greatly reduced, thereby achieving the advantages of high efficiency and high detection accuracy.

[0118] In a further configuration of the present invention, the canny edge detection is used to detect the image g σ (x, y) for edge detection, the specific method is as follows:

[0119] First, calculate the gradient value and gradient direction of the image, and calculate the gradient value G(x, y) and gradient direction θ by the following formula:

[0120]

[0121]

[0122] where g x (x, y) 2 , g y (x, y) 2 Respectively represent g σ (x, y) returns the derivative values ​​in the horizontal and vertical directions through Sobel;

[0123] The gradient value G(x, y) is non-maximum suppressed according to the following formula:

[0124]

[0125] Where T1 represents the threshold of non-maximum suppression, filtering out noise;

[0126] According to the following formula, G T1 (x, y) performs double threshold screening:

[0127]

[0128] Where maxVal represents the high threshold, minVal represents the low threshold, and edge represents the pixel of the image to be retained, otherwise it is cleared. canny (x, y) represents the edge image after filtering.

[0129] According to a further configuration of the present invention, the method for determining whether the front half of the bobbin contains yarn is as follows:

[0130] According to the following formula, the edge image G canny (x, y) performs gradient statistics and performs threshold judgment to determine whether the first half contains yarn:

[0131]

[0132] Among them G canny .width indicates edge image G canny The width of (x, y),

[0133] If there exists j1 such that If T2 is the threshold of whether the first half contains yarn, then the first half of the bobbin contains yarn, otherwise the first half of the bobbin does not contain yarn, where j1∈(0, len(h(y))-α) represents the starting coordinates of the sliding summation window, α1 represents the length of the sliding summation window, and len(h(y)) represents the length of h(y).

[0134] Wherein, the method for extracting the color of the yarn tube is as follows:

[0135] For the filtered bobbin image g σ (x, y) is grayed out according to the following formula:

[0136] Gray=R*0.299+G*0.587+B*0.114

[0137]

[0138] Gray represents the grayscale formula. σ Each pixel in (x, y) is grayed out according to the formula to obtain g Gray (x, y);

[0139] Identify the reflection and dirty spots of the yarn tube;

[0140] Extract the tube body color.

[0141] Specifically, the method for judging the reflection and dirty spots of the yarn tube is as follows:

[0142] Use the following formula to determine whether there are reflective points and extract the grayscale values ​​of reflective points and dirty points:

[0143] Ref=MAX(g Gray (x, y)

[0144] Dir=MIN(g Gray (x, y)

[0145]

[0146] Where Ref represents the gray value of the reflective point, Dir represents the gray value of the dirty point, and T3 represents the threshold value of whether there is a reflective point;

[0147] If the grayscale value of the reflective point and the grayscale value of the dirty point are greater than the threshold T3, it means there is a reflective point, otherwise there is no reflective point.

[0148] Furthermore, the method for extracting the tube body color is as follows:

[0149] After separating the RGB channels, g bσ , g gσ , g rσ Each pixel point (x∈(0, g σ .width), y∈(0, g σ .height / 2),) color extraction is performed according to the following formula:

[0150] If there are reflective spots:

[0151]

[0152] If there is no reflective point

[0153]

[0154] The R component of the tube body color is MAX(R_NUM), where R_NUM represents the number of pixels of the R color component between 0 and 255. σ .width indicates the width of the image, g σ .height indicates the length of the image, T4 indicates the threshold of whether a pixel is valid, T5 indicates the threshold of whether a pixel is a reflective point, and the B and G components are operated according to the above operation. Similarly, the RGB color value of the tube body can be obtained.

[0155] Further, the method for judging whether the rear half of the bobbin contains yarn is as follows:

[0156] Match the pixel points in each column of the second half of the tube with the RGB value of the tube body and accumulate them:

[0157]

[0158] Where T6 represents the threshold value of whether the pixel point successfully matches the tube body RGB, and Mat(y) represents the number of pixels in each column that match the tube body;

[0159] Mat(y) is accumulated using a sliding window according to the following formula:

[0160]

[0161] Where j2∈(0, len(Mat)-α2) represents the starting coordinates of the sliding window, α2 represents the length of the sliding window, T7 represents the threshold of the sliding window to determine whether there is yarn, len(Mat) represents the length of the matching array Mat(y), slid(j2) represents the accumulated value of Mat(y) after passing through the sliding window, if the accumulated value is greater than the threshold T7, it means there may be yarn,

[0162] The residual yarn detection of the yarn-containing tube is carried out by the following formula:

[0163]

[0164] Where len(slid) represents the length of slid, and T8 represents the threshold for judging whether there is yarn;

[0165] If the number of times the possible yarn appears is greater than the threshold value T8, it is determined that the bobbin contains yarn, otherwise it does not contain yarn, and the determination of whether the bobbin contains yarn is completed.

[0166] In the above technical solution, the yarn tube is split into two parts, the front and rear parts, and identification and judgment are performed separately, thereby improving the detection accuracy.

[0167] In addition, by extracting the color of the tube body, the reflection and dirty spots are judged through color comparison, and then a secondary judgment is completed to further ensure the detection accuracy of yarn residue.

[0168] Example 2

[0169] This embodiment discloses a yarn residue detection device for a yarn management machine, including a memory, a processor, and a program stored in the memory, wherein the program is configured to be executed by the processor, and when the processor executes the program, the yarn residue detection method for the yarn management machine described in Example 1 is implemented.

[0170] Example 3

[0171] This embodiment discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the yarn residue detection method of the yarn management machine described in Embodiment 1 is implemented.

[0172] The embodiments of the present invention are described in detail above, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principles and spirit of the present invention, various changes, modifications, substitutions and variations of these embodiments including components still fall within the scope of protection of the present invention.

Claims

1. A yarn residue detection method for a yarn management machine, characterized in that: The steps include: Step 1: Collect the RGB image of the current bobbin The interception of the RGB image is recorded as f, f(left, top, right, bottom) represents the interception of the collected f, left, top, right, bottom represent the values ​​of the left, top, right, and bottom of f respectively, let g(x, y) = f(left, top, right, bottom) represent the intercepted bobbin image, where x and y represent the ordinate and abscissa of the image respectively; Step 2: Preprocess the image Step 2.1, perform Gaussian filtering on the image; Step 2.2, separate the RGB channels; Step 2.3, perform Gaussian filtering on the separated RGB channel image; Step 2.4, perform canny edge detection on the filtered image; The canny edge detection is used to detect the image g σ (x, y) for edge detection. The specific method is as follows: First, calculate the gradient value and gradient direction of the image, and calculate the gradient value G(x, y) and gradient direction θ using the following formula: where g x (x,y) 2 , g y (x,y) 2 Respectively represent g σ (x,y) returns the derivative values ​​in the horizontal and vertical directions through Sobel; The gradient value G(x,y) is non-maximum suppressed according to the following formula: Where T1 represents the threshold of non-maximum suppression, filtering out noise; According to the following formula, G T1 (x,y) performs double threshold screening: Where maxVal represents the high threshold, minVal represents the low threshold, and edge represents the pixel of the image to be retained, otherwise it is cleared. canny (x,y) represents the edge image after filtering; Step 3, detecting the residual yarn on the bobbin; Step 3.1, determine whether the front half of the bobbin contains yarn; Step 3.2, extracting the color of the yarn tube; The method for extracting the color of the yarn tube is as follows: For the filtered bobbin image g σ (x,y) is grayed out according to the following formula: Gray=R*0.299+G*0.587+B*0.114 Gray represents the grayscale formula. σ Each pixel in (x, y) is grayed out according to the formula to obtain g Gray (x,y); Judging the reflection and dirt of the yarn tube: Use the following formula to determine whether there are reflective points and extract the grayscale values ​​of reflective points and dirty points: Ref=MAX(g Gray (x,y)) Dir=MIN(g Gray (x,y)) Where Ref represents the gray value of the reflective point, Dir represents the gray value of the dirty point, and T3 represents the threshold value of whether there is a reflective point; If the grayscale value of the reflective point and the grayscale value of the dirty point are greater than the threshold T3, it means there is a reflective point, otherwise there is no reflective point; Extract the tube body color: After separating the RGB channels, g bσ , g gσ , g rσ For each pixel point (x∈(0,g σ .width),y∈(0,g σ .height / 2)) Color extraction is performed according to the following formula: If there are reflective spots: If there is no reflective point The R component of the tube body color is MAX(R_NUM), where R_NUM represents the number of pixels of the R color component between 0 and 255. σ .width indicates the width of the image, g σ .height indicates the length of the image, T4 indicates the threshold value of whether a pixel is valid, T5 indicates the threshold value of whether a pixel is a reflective point, and the B and G components are operated according to the above operation, and the RGB color value of the tube body can be obtained in the same way; Step 3.3: Determine whether the second half of the bobbin contains yarn: Match the pixel points in each column of the second half of the tube with the RGB value of the tube body and accumulate them: Where T6 represents the threshold value of whether the pixel point successfully matches the tube body RGB, and Mat(y) represents the number of pixels in each column that match the tube body; Mat(y) is accumulated using a sliding window according to the following formula: Where j2∈(0,len(Mat)-α2) represents the starting coordinates of the sliding window, α2 represents the length of the sliding window, T7 represents the threshold of the sliding window to determine whether there is yarn, len(Mat) represents the length of the matching array Mat(y), slid(j2) represents the accumulated value of Mat(y) after passing through the sliding window, if the accumulated value is greater than the threshold T7, it means there may be yarn, The residual yarn detection of the yarn-containing tube is carried out by the following formula: Where len(slid) represents the length of slid, and T8 represents the threshold for judging whether there is yarn; If the number of times the possible yarn appears is greater than the threshold value T8, it is determined that the bobbin contains yarn, otherwise it does not contain yarn, and the determination of whether the bobbin contains yarn is completed.

2. The yarn residue detection method of the yarn management bobbin machine according to claim 1 is characterized in that: In step 2, the intercepted bobbin image g(x,y) is preprocessed. In step 2.1, Gaussian filtering is performed on g(x,y) according to the following formula: g σ (x,y)=g(x,y)*G σ Where * represents the convolution operation, G σ is a two-dimensional Gaussian kernel with standard deviation σ, defined as: In step 2.2, g is calculated according to the following formula: σ (x,y) performs RGB channel separation: g b =g σ [0] g g =g σ [1] g r =g σ [2] where g b Indicates the B channel of the bobbin image, g g Represents the G channel of the bobbin image, g r The R channel represents the bobbin image; In step 2.3, repeat step 2.1 to perform Gaussian filtering on the separated RGB channel images respectively: g bσ =g b *G σ g gσ =g g *G σ g rσ =g r *G σ 。 3. The yarn residue detection method of the yarn management bobbin machine according to claim 2 is characterized in that: The method for judging whether the front half of the bobbin contains yarn is as follows: According to the following formula, the edge image G canny (x, y) performs gradient statistics and performs threshold judgment to determine whether the first half contains yarn: Among them G canny .width indicates edge image G canny The width of (x,y), If there exists j1 such that Then the first half of the bobbin contains yarn, otherwise the first half of the bobbin does not contain yarn, where T2 represents the threshold of whether the first half contains yarn, j1∈(0,len(h(y))-α) represents the starting coordinates of the sliding summation window, α1 represents the length of the sliding summation window, and len(h(y)) represents the length of h(y).

4. A yarn residue detection device for a yarn management machine, characterized in that: The invention comprises a memory, a processor and a program stored in the memory, wherein the program is configured to be executed by the processor, and the yarn residue detection method of the yarn management machine according to any one of claims 1 to 3 is implemented when the processor executes the program.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the yarn residue detection method of the yarn management bobbin machine according to any one of claims 1 to 3 is implemented.

Images