A method, device, equipment and storage medium for detecting broken yarn of white bundled yarn

Abstract

The present invention relates to the field of glass fiber detection technology, and in particular to a method, device, equipment and storage medium for detecting broken yarns of white bundled yarns, comprising the following steps: setting a monitoring area and setting a screening angle value, a screening height value, a spacing value and a sheet value, obtaining color image data and converting it into grayscale image data; filtering using mean curvature flow, and outputting image data to be processed; extracting the outline of white lines from the image data to be processed using a line extractor, and outputting sub-pixel precision lines; screening the sub-pixel precision lines, and sorting them by column coordinates, outputting the result outline and performing straight line fitting; judging the distance between the fitted straight lines, and considering it as broken yarn if it exceeds the set spacing value; considering it as broken yarn if the distance between the first yarn and the last yarn and the reference line exceeds the set spacing value; and alarming when the broken yarn image is continuously processed and the set sheet value is set. The present invention performs broken yarn and parallel yarn detection on the bundled yarn on the glass fiber cloth by visual detection.

Description

Technical Field

[0001] The invention relates to the technical field of glass fiber detection, and in particular to a method, a device, equipment and a storage medium for detecting broken white bundled yarns. Background Art

[0002] Glass fiber cloth is mainly used in industry for heat insulation, fire prevention and flame retardancy. When it is burned by flames, the material absorbs a lot of heat and can prevent the flame from passing through and isolate the air. The axial fabric of glass fiber cloth is made by weaving glass fiber direct untwisted roving on a multi-axial warp knitting machine. The warp and weft yarns are arranged in one or more layers according to the preset angles and then bound with binding yarns. The binding yarns are mostly white yarns.

[0003] In the process of mass production of glass fiber fabric, warp knitting machines may experience yarn breakage and doubling of white bundled yarns, which affects product quality and requires machine vision to detect such defects. In the prior art, small plane models are often used to detect yarn breakage and doubling of bundled yarns. On the one hand, the parameters are difficult to adjust, and on the other hand, there may be situations where the lines cannot be detected, which is not suitable for application in large-scale production lines.

[0004] The detection of broken yarn and doubling of bundled yarn in large-scale glass fiber cloth production lines has become a technical problem that needs to be solved urgently.

[0005] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] In view of at least one of the above technical problems, the present invention provides a method, device, equipment and storage medium for detecting broken yarns in white bundled yarns, which use a visual method to detect broken yarns and doubling on the glass fiber cloth surface.

[0007] According to a first aspect of the present invention, a method for detecting broken white bundled yarn is provided, comprising the following steps:

[0008] S10: Setting the monitoring area and setting the screening angle value, the screening height value, the spacing value and the number of sheets to obtain the color image data of the glass fiber cloth surface;

[0009] S20: Convert the color image data into grayscale image data;

[0010] S30: filtering the grayscale image data using mean curvature flow, and outputting the filtered image data as image data to be processed;

[0011] S40: extracting the outline of the white line from the image data to be processed by a line extractor, and outputting the outline as a sub-pixel precision line;

[0012] S50: filtering the sub-pixel precision lines by using the set filtering angle value and the set filtering height value, and sorting them from left to right according to column coordinates, and outputting them as result contours;

[0013] S60: performing straight line fitting on the result profile;

[0014] S70: Determine the distance between the two fitting straight lines. If the distance exceeds the set distance value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set distance value, it is also considered as a broken yarn.

[0015] S80: When the yarn-breaking image is detected to continuously process the set number of sheets, an alarm is issued.

[0016] In some embodiments of the present invention, in step S30, the grayscale image data is filtered using the mean curvature flow as follows:

[0017] The curvature flow equation is:

[0018]

[0019] Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

[0020] In some embodiments of the present invention, step S40 further includes the following steps:

[0021] S41: Set the Gaussian filter value to 1, set the high value of the hysteresis threshold segmentation to 4, and set the low value of the hysteresis threshold segmentation to 3;

[0022] S42: Calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel;

[0023] S43: Select the local maximum pixel value in the second-order partial derivative perpendicular to the line direction and mark it as a line point;

[0024] S44: if the second-order partial derivative of the line point is greater than the setting value of the hysteresis threshold segmentation high value, the first pre-selected point is output; if the second-order partial derivative of the line point is less than the setting value of the hysteresis threshold segmentation low value, the line point is discarded;

[0025] S45: if other line points are connected to the first pre-selected point through a connection path, then output as a second pre-selected point;

[0026] S46: Combining the first pre-selected point with the second pre-selected point and outputting them as a sub-pixel precision line.

[0027] In some embodiments of the present invention, in step S60, the straight line fitting method includes the following steps:

[0028] S61: Set the outlier elimination factor to 2, and set the maximum number of iterations N to 5;

[0029] S62: According to the following normal equation:

[0030] r·N r +c·N c -Dist=0,

[0031] Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation;

[0032] S63: Fitting straight lines to the pixels in the result contour using the least squares method, the formula is as follows:

[0033]

[0034] S64: Assigning weights to the pixel points in the result contour, wherein the weights of the pixel points far from the straight line are smaller than the weights of the pixel points close to the straight line.

[0035] According to a second aspect of the present invention, there is also provided a white bundled yarn breakage detection device, comprising:

[0036] Acquisition module: used to set the monitoring area and set the screening angle value, screening height value, spacing value and sheet value to obtain the color image data of the glass fiber cloth surface;

[0037] Image processing module: used for converting the color image data into grayscale image data;

[0038] Filtering module: used for filtering the grayscale image data using mean curvature flow, and outputting the grayscale image data as image data to be processed;

[0039] Extraction module: used for extracting the outline of the white line from the image data to be processed by using a line extractor, and outputting the outline as a sub-pixel precision line;

[0040] Screening module: used for screening the sub-pixel precision lines by the set screening angle value and the set screening height value, and sorting them from left to right according to the column coordinates, and outputting the result outline;

[0041] Fitting module: used for performing straight line fitting on the result profile;

[0042] Judgment module: used to judge the distance between the two fitting lines. If the distance exceeds the set distance value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set distance value, it is also considered as a broken yarn.

[0043] Alarm module: used for giving an alarm when the yarn-breaking image is detected to continuously process the set number of sheets.

[0044] In some embodiments of the present invention, in the filtering module, the grayscale image data is filtered using the mean curvature flow as follows:

[0045] The curvature flow equation is:

[0046]

[0047] Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

[0048] In some embodiments of the present invention, the extraction module further includes the following units:

[0049] Setting unit: used to set the Gaussian filter value to 1, the hysteresis threshold segmentation high value to 4, and the hysteresis threshold segmentation low value to 3;

[0050] Computation unit: used to calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel;

[0051] A selection unit: used for selecting a local maximum pixel value in the second-order partial derivative perpendicular to the line direction, and marking it as a line point;

[0052] A first screening unit: for outputting a first pre-selected point if the second-order partial derivative of the line point is greater than a setting value of a hysteresis threshold segmentation high value, and for discarding the line point if the second-order partial derivative of the line point is less than a setting value of a hysteresis threshold segmentation low value;

[0053] A second screening unit: configured to output other line points as second pre-selected points if they are connected to the first pre-selected points through a connection path;

[0054] Output unit: used for combining the first pre-selected point with the second pre-selected point and outputting them as a sub-pixel precision line.

[0055] In some embodiments of the present invention, in the fitting module, the straight line fitting method includes the following units:

[0056] Setting unit: used to set the outlier elimination factor to 2 and the maximum number of iterations N to 5;

[0057] Get unit: Used according to the following normal equation:

[0058] r·N r 10c·N c -Dist=0,

[0059] Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation;

[0060] Fitting unit: used to fit a straight line to the pixel points in the result contour using the least squares method, the formula is as follows:

[0061]

[0062] Weighting unit: used to assign weights to the pixel points in the result contour, and the weights of the pixel points far from the straight line are smaller than the weights of the pixel points close to the straight line.

[0063] According to a third aspect of the present invention, there is also provided a white bundled yarn breakage detection device, comprising a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described above is implemented.

[0064] According to a fourth aspect of the present invention, there is also provided a white bundled yarn breakage detection storage medium, comprising a storage medium on which a computer program is stored, and the computer program implements the above method when executed by a processor.

[0065] The beneficial effects of the present invention are as follows: the present invention obtains color image data by photographing and sampling the glass fiber cloth surface, converts it into grayscale image data, processes the grayscale image data, filters the grayscale image data using mean curvature flow, and uses a line extractor to extract the outline of white lines in the grayscale image data after filtering, and outputs it as a sub-pixel longitude line, filters the sub-pixel longitude line according to the set screening angle value and screening height value, and sorts it from left to right according to the column coordinates, and performs straight line fitting, and judges the straight line after fitting, and judges the distance between two adjacent fitting straight lines. If it exceeds the set spacing value, it is regarded as broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set spacing value, it is also regarded as broken yarn. If the image of the detected broken yarn continuously processes the set number of sheets, an alarm is issued. The white bundled yarn is judged by visual recognition, and the broken yarn and the doubling situation are shortened from the detection of broken yarn to the shutdown time, and the production cost caused by the broken yarn is reduced to improve the quality of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0067] Figure 1 This is a flowchart of the steps of the method for detecting broken white bundled yarn in an embodiment of the present invention;

[0068] Figure 2 2 is a schematic structural diagram of a method for detecting broken white bundled yarn in an embodiment of the present invention;

[0069] Figure 3 Schematic diagram of the structure of the device of the method for detecting broken white bundled yarn in an embodiment of the present invention. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0071] like Figures 1 to 3 A method for detecting broken yarn of white bundled yarn is shown, comprising the following steps:

[0072] S10: Set the monitoring area and set the screening angle value, set the screening height value, set the spacing value and set the sheet value to obtain the color image data of the glass fiber cloth surface. When performing yarn breakage and yarn doubling detection on the glass fiber cloth surface, first set the monitoring area and determine the values ​​of various parameters, and then obtain the color image data of the glass fiber cloth surface.

[0073] S20: converting the color image data into grayscale image data, and then calculating the grayscale image data to detect the broken yarn and the doubling of the white bundled yarn on the glass fiber cloth. It should be pointed out here that there are many ways to convert the color image data into grayscale image data, which can be an average method, a maximum and minimum average method, a weighted average method or other conversion methods.

[0074] S30: filtering the grayscale image data using the mean curvature flow, and outputting the grayscale image data as the image data to be processed. Filtering the grayscale image data using the mean curvature flow can make the image smoother. It should be noted here that there are many ways to filter the image, which can be box filtering, mean filtering, Gaussian filtering or other filtering methods.

[0075] S40: Use a line extractor to extract the outline of the white lines from the image data to be processed, and output it as a sub-pixel precision line. The white binding yarn on the glass fiber cloth is displayed on the grayscale image data. Use a line extractor to extract the outline of the white lines on the image data to be processed. The sub-pixel precision line has a higher precision than the pixel. The coordinates of the pixel are positive integers, and the coordinates of the sub-pixel precision line can be accurate to decimal places.

[0076] S50: Sub-pixel precision lines are screened by setting a screening angle value and a screening height value, and are sorted from left to right according to the column coordinates, and the result outline is output. The extracted sub-pixel precision lines contain white bundled yarns and some broken lines and oblique lines. Through height screening and angle screening, the broken lines and oblique lines are screened out to reduce interference items in the judgment of white lines. It should be pointed out here that the angle screening can be set to -1.6 radians to -1.2 radians and 1.2 radians to 1.6 radians, and the height screening is set to 30 pixels, which can also be changed according to actual conditions.

[0077] S60: Linear fitting is performed on the result contour. Linear fitting is mainly to create a straight line that is relatively close to the original contour line. By performing linear fitting on the result contour, the sub-pixel precision line is synthesized into a line segment, and then the line segment is judged. It should be pointed out here that there are many ways of linear fitting, which can be the least squares method, the gradient descent method, or the Gauss-Newton method.

[0078] S70: Determine the distance between the two fitting straight lines. If it exceeds the set spacing value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set spacing value, it is also considered as a broken yarn. The yarn breakage is determined by the set spacing value. When the distance between two adjacent straight lines does not meet the set spacing value, it is considered as a broken yarn. The yarn breakage of the first yarn and the last yarn is determined by the reference line. If the distance between the first yarn or the last yarn and the reference line does not meet the set spacing value, it is also determined as a broken yarn. It should be pointed out here that the set spacing value can be set according to the distance between the white bundled yarns on the glass fiber cloth.

[0079] S80: When the detected image of broken yarn is processed continuously for a set number of sheets, an alarm is issued, and whether the yarn is broken is determined by the set number of sheets. When the detected image continuously processes the set number of sheets and a yarn break occurs, it is regarded as a yarn break, and an alarm is issued to shut down the machine. After the shutdown is completed, the distance from the place where the yarn break begins to the needle head of the warp knitting machine shall not exceed 30 cm. The glass fiber cloth surface is judged through visual recognition, which shortens the time from detecting the yarn break to shutting down, and reduces the production cost caused by the yarn break.

[0080] like Figure 1 As shown, the present invention obtains color image data by photographing and sampling the glass fiber cloth surface, converts it into grayscale image data, processes the grayscale image data, filters the grayscale image data using mean curvature flow, and uses a line extractor to extract the outline of the white line in the grayscale image data after filtering, and outputs it as a sub-pixel longitude line, screens the sub-pixel longitude line according to the set screening angle value and screening height value, and sorts it from left to right according to the column coordinates, and performs straight line fitting, and judges the fitted straight line, and judges the distance between two adjacent fitting straight lines. If it exceeds the set spacing value, it is regarded as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set spacing value, it is also regarded as a broken yarn. If the image of the detected broken yarn continuously processes the set number of sheets, an alarm is issued. The white bundled yarn is judged by visual recognition, and the broken yarn and the doubling situation are shortened from the detection of the broken yarn to the shutdown, and the production cost caused by the broken yarn is reduced to improve the quality of the product.

[0081] In some embodiments of the present invention, in step S30, the grayscale image data is filtered using the mean curvature flow as follows:

[0082] The curvature flow equation is:

[0083]

[0084] Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

[0085] The mean curvature flow filtering is performed along the edge. In some embodiments of the present invention, the grayscale image is filtered by the above equation. This is mainly because in some embodiments of the present invention, the glass fiber cloth is photographed by a camera. When the color image data is output, it has been processed by the camera. The color image data has less noise and can be directly filtered by the mean curvature flow.

[0086] In some embodiments of the present invention, step S40 further includes the following steps:

[0087] S41: Set the Gaussian filter value to 1, set the high value of the hysteresis threshold segmentation to 4, and set the low value of the hysteresis threshold segmentation to 3;

[0088] S42: Calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel;

[0089] S43: Select the local maximum pixel value in the second-order partial derivative perpendicular to the line direction and mark it as a line point;

[0090] S44: if the second-order partial derivative of the line point is greater than the setting value of the hysteresis threshold segmentation high value, the output is the first pre-selected point; if the second-order partial derivative of the line point is less than the setting value of the hysteresis threshold segmentation low value, the line point is discarded;

[0091] S45: If other line points are connected to the first pre-selected point through the connection path, then the second pre-selected point is output;

[0092] S46: Combining the first pre-selected point with the second pre-selected point and outputting them as a sub-pixel precision line.

[0093] In some embodiments of the present invention, a line extractor is used to detect lines and their widths. It should be pointed out here that the parameter Gaussian filter value determines the degree of image smoothing. The parameter values ​​are set according to the actual situation. First, the partial derivatives of the Gaussian filter mask and the image convolution kernel are calculated to determine the line direction of the parameters and pixels. Then, the local maximum pixel value in the second-order partial derivative perpendicular to the line direction is selected and marked as a line point. Finally, the line point is judged to determine the size between the second-order partial derivative of the line point and the set value of the hysteresis threshold segmentation high value and the hysteresis threshold segmentation low value. It is chosen to discard or retain. When the retained first pre-selected point is connected, there are also some second pre-selected points that can be connected to it. The first pre-selected point and the second pre-selected point are connected together, and finally output as a sub-pixel precision line. The larger the Gaussian filter mask, the deeper the image smoothing.

[0094] In some embodiments of the present invention, in step S60, the straight line fitting method includes the following steps:

[0095] S61: Set the outlier elimination factor to 2, and set the maximum number of iterations N to 5;

[0096] S62: According to the following normal equation:

[0097] r·N r 10c·N c -Dist=0,

[0098] Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation;

[0099] S63: Use the least squares method to fit a straight line to the pixel points in the result contour. The formula is as follows:

[0100]

[0101] S64: Assigning weights to the pixel points in the result contour, wherein the weights of the pixel points far from the straight line are smaller than the weights of the pixel points close to the straight line.

[0102] In some embodiments of the present invention, various parameters are first set according to actual conditions, the coordinates of the starting point of the line segment and the vectors of Nr and Nc in the normal equation are obtained through the normal equation, and then the least squares method is used to fit the pixels in the resulting contour to a straight line. Finally, the pixels in the resulting contour are weighted to ensure the accuracy of the fitting. It should be pointed out here that in order to further improve the accuracy of the fitting, the outlier rejection factor is set to 2, and the maximum number of iterations N is set to 5, which can also be changed according to actual conditions.

[0103] Those skilled in the art should know that the embodiments of the present application can be provided as methods, devices, electronic devices or computer storage media products. Therefore, the embodiments of the present application can completely adopt hardware embodiments, embodiments combining hardware and software, or pure software embodiments. The test detection process data processing device in the embodiments of the present application is introduced below. The device embodiments below correspond to the method embodiments above. Those skilled in the art can understand the implementation process below based on the above description, and no detailed description is given here.

[0104] like Figure 3 The white bundled yarn breakage detection device shown in the figure comprises:

[0105] Acquisition module: used to set the monitoring area and set the screening angle value, screening height value, spacing value and sheet value to obtain the color image data of the glass fiber cloth surface;

[0106] Image processing module: used to convert color image data into grayscale image data;

[0107] Filtering module: used to filter the grayscale image data using the mean curvature flow, and output it as the image data to be processed;

[0108] Extraction module: used to extract the outline of white lines from the image data to be processed using a line extractor, and output it as a sub-pixel precision line;

[0109] Filtering module: used to filter sub-pixel precision lines by setting the filtering angle value and the filtering height value, and sort them from left to right according to the column coordinates, and output the result outline;

[0110] Fitting module: used to perform straight line fitting on the result contour;

[0111] Judgment module: used to judge the distance between the two fitting lines. If it exceeds the set distance value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set distance value, it is also considered as a broken yarn.

[0112] Alarm module: used to alarm when the yarn-breaking image is detected and processed continuously for a set number of sheets.

[0113] In some embodiments of the present invention, in the filtering module, the grayscale image data is filtered using the mean curvature flow as follows:

[0114] The curvature flow equation is:

[0115]

[0116] Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

[0117] In some embodiments of the present invention, the re-extraction module further includes the following units:

[0118] Setting unit: used to set the Gaussian filter value to 1, the hysteresis threshold segmentation high value to 4, and the hysteresis threshold segmentation low value to 3;

[0119] Computation unit: used to calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel;

[0120] Selection unit: used to select the local maximum pixel value in the second-order partial derivative perpendicular to the line direction and mark it as a line point;

[0121] The first screening unit is used to output the first pre-selected point if the second-order partial derivative of the line point is greater than the setting value of the hysteresis threshold segmentation high value, and to discard the line point if the second-order partial derivative of the line point is less than the setting value of the hysteresis threshold segmentation low value;

[0122] A second screening unit: used for outputting other line points as second pre-selected points if they are connected to the first pre-selected points through a connection path;

[0123] Output unit: used to combine the first pre-selected point with the second pre-selected point and output them as a sub-pixel precision line.

[0124] In some embodiments of the present invention, in the fitting module, the straight line fitting method includes the following units:

[0125] Setting unit: used to set the outlier elimination factor to 2 and the maximum number of iterations N to 5;

[0126] Get unit: Used according to the following normal equation:

[0127] r·N r 10c·N c -Dist=0,

[0128] Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation;

[0129] Fitting unit: used to fit straight lines to the pixels in the result contour using the least squares method. The formula is as follows:

[0130]

[0131] Weighting unit: used to assign weights to the pixels in the resulting contour. The weights of pixels far from the straight line are smaller than those of pixels close to the straight line.

[0132] In the following part of the embodiments of the present invention, the electronic device and computer storage medium embodiments in the embodiments of the present invention are introduced. The electronic device and computer storage medium embodiments in the following correspond to the method embodiments in the above. Those skilled in the art can understand the following implementation process based on the above description, which will not be described in detail here.

[0133] like Figure 2 The structural diagram of the computer device provided in the embodiment of the present application is shown. A computer device 400 provided in the embodiment of the present application includes: a memory 420, a processor 410, and a computer program stored in the memory 420 and executable on the processor 410. When the processor 410 executes the computer program, the above method is implemented.

[0134] The embodiment of the present application further provides a storage medium 430 on which a computer program is stored. When the computer program is executed by the processor 410, the above method is implemented.

[0135] Among them, the storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0136] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0137] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0138] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0139] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0140] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can obtain instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0141] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0142] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0143] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for detecting broken white bundled yarn, characterized in that: The steps include: S10: Setting the monitoring area and setting the screening angle value, the screening height value, the spacing value and the number of sheets to obtain the color image data of the glass fiber cloth surface; S20: Convert the color image data into grayscale image data; S30: filtering the grayscale image data using mean curvature flow, and outputting the filtered image data as image data to be processed; S40: extracting the outline of the white line from the image data to be processed by a line extractor, and outputting the outline as a sub-pixel precision line; S50: filtering the sub-pixel precision lines by using the set filtering angle value and the set filtering height value, and sorting them from left to right according to column coordinates, and outputting them as result contours; S60: performing straight line fitting on the result profile; S70: Determine the distance between the two fitting straight lines. If the distance exceeds the set distance value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set distance value, it is also considered as a broken yarn. S80: when the yarn-breaking image is detected and the set number of sheets is processed continuously, an alarm is issued; In step S40, the following steps are also included: S41: Set the Gaussian filter value to 1, set the high value of the hysteresis threshold segmentation to 4, and set the low value of the hysteresis threshold segmentation to 3; S42: Calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel; S43: Select the local maximum pixel value in the second-order partial derivative perpendicular to the line direction and mark it as a line point; S44: if the second-order partial derivative of the line point is greater than the setting value of the hysteresis threshold segmentation high value, the first pre-selected point is output; if the second-order partial derivative of the line point is less than the setting value of the hysteresis threshold segmentation low value, the line point is discarded; S45: if other line points are connected to the first pre-selected point through a connection path, then output as a second pre-selected point; S46: Combine the first pre-selected point and the second pre-selected point, and output them as a sub-pixel precision line.

2. The method for detecting broken white bundled yarn according to claim 1, characterized in that: In step S30, the grayscale image data is filtered using the mean curvature flow as follows: The curvature flow equation is: ； Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

3. The method for detecting broken white bundled yarn according to claim 1, characterized in that: In step S60, the straight line fitting method includes the following steps: S61: Set the outlier elimination factor to 2, and set the maximum number of iterations N to 5; S62: According to the following normal equation: ； Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation; S63: Fitting straight lines to the pixels in the result contour using the least squares method, the formula is as follows: ； S64: Assigning weights to the pixel points in the result contour, wherein the weights of the pixel points far from the straight line are smaller than the weights of the pixel points close to the straight line.

4. A white bundled yarn breakage detection device, characterized in that: include: Acquisition module: used to set the monitoring area and set the screening angle value, screening height value, spacing value and sheet value to obtain the color image data of the glass fiber cloth surface; Image processing module: used for converting the color image data into grayscale image data; Filtering module: used for filtering the grayscale image data using mean curvature flow, and outputting the grayscale image data as image data to be processed; Extraction module: used for extracting the outline of the white line from the image data to be processed by using a line extractor, and outputting the outline as a sub-pixel precision line; Screening module: used for screening the sub-pixel precision lines by the set screening angle value and the set screening height value, and sorting them from left to right according to the column coordinates, and outputting the result outline; Fitting module: used for performing straight line fitting on the result profile; Judgment module: used to judge the distance between the two fitting lines. If the distance exceeds the set distance value, it is considered as a broken yarn. If the distance between the first yarn or the last yarn and the reference line exceeds the set distance value, it is also considered as a broken yarn. Alarm module: used for alarming when the yarn-breaking image is detected to continuously process the set number of sheets; The extraction module further includes the following units: Setting unit: used to set the Gaussian filter value to 1, the hysteresis threshold segmentation high value to 4, and the hysteresis threshold segmentation low value to 3; Computation unit: used to calculate the partial derivatives of the Gaussian filter mask and the image convolution kernel, determine the quadratic polynomial parameters of each pixel point on the image in the x and y directions, and calculate the line direction of each pixel; A selection unit: used for selecting a local maximum pixel value in the second-order partial derivative perpendicular to the line direction, and marking it as a line point; A first screening unit: for outputting a first pre-selected point if the second-order partial derivative of the line point is greater than a setting value of a hysteresis threshold segmentation high value, and for discarding the line point if the second-order partial derivative of the line point is less than a setting value of a hysteresis threshold segmentation low value; A second screening unit: configured to output other line points as second pre-selected points if they are connected to the first pre-selected points through a connection path; Output unit: used to combine the first pre-selected point with the second pre-selected point and output them as a sub-pixel precision line.

5. The white bundled yarn breakage detection device according to claim 4, characterized in that: In the filtering module, the grayscale image data is filtered using the mean curvature flow as follows: The curvature flow equation is: ； Among them, u is the gray value function, and the time step T is set to 0.5, the number of iterations N is 10, the parameter sigma value is taken as 0.5, and the calculation result is the image data to be processed containing the gray value function at time T*N.

6. The white binding yarn according to claim 4 The yarn breakage detection device is characterized in that: In the fitting module, the straight line fitting method includes the following units: Setting unit: used to set the outlier elimination factor to 2 and the maximum number of iterations N to 5; Get unit: Used according to the following normal equation: ； Get the coordinates of the starting point of the line segment, as well as the vectors Nr and Nc in the normal equation; Fitting unit: used to fit a straight line to the pixel points in the result contour using the least squares method, the formula is as follows: ； Weighting unit: used to assign weights to the pixel points in the result contour, and the weights of the pixel points far from the straight line are smaller than the weights of the pixel points close to the straight line.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 3 is implemented.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images