Forming balance detection method for wet paper web material of toilet paper

Abstract

The invention belongs to the technical field of intelligent detection and machine vision, and provides a forming balance detection method for a wet paper web material of toilet paper, which specifically comprises the following steps: arranging a CCD (Charge Coupled Device) camera on a wet paper web conveying belt, performing real-time data acquisition through the CCD camera to obtain a paper web image, preprocessing the paper web image, and determining the forming balance of the wet paper web material of the toilet paper; and then counting the number of uniform failure points and a failure point estimation range from the paper web image, carrying out speed matching decision analysis through the number of the failure points and the failure point estimation range to form a decision calculation value, and finally judging the uniformity of the current wet paper web by utilizing the decision calculation value. According to the method, the risk of functional decline occurring in the process of uniformly distributing the pulp by using the forming mesh machine due to too high forming speed of the wet paper web in the production process is identified, and the wet paper web with abnormal paper flatness and surface smoothness quality caused by uneven stress is found in time; the distribution uniformity of the toilet paper material in the wet paper web process is accurately evaluated, and the loss of raw materials and the production cost are remarkably reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent detection and machine vision, and in particular relates to a method for detecting the forming balance of a wet paper web material of toilet paper. Background Art

[0002] The flatness of paper is a crucial indicator in the material testing of toilet paper. On the one hand, its flatness reflects the good uniformity of paper material distribution, and on the other hand, it better ensures comfort during use. In high-end toilet paper products, the risk of damage caused by uneven thickness will be more concerned. Usually, in the production process of toilet paper, the relevant testing of paper flatness is usually based on the finished product, but this test will result in a large amount of raw material loss. The principle that causes the flatness of paper is that there is always a dynamic matching decision range between the forming speed of the wet paper web and the uniform spreading quality of the slurry, and the forming speed of the wet paper web is usually preset according to the material of the toilet paper and the forming mesh cloth equipment, so the forming speed of the wet paper web can guarantee a relatively stable output effect within the matching decision range, but the production system always has a high tendency for work efficiency, and on this basis, a certain degree of abnormality is allowed to occur. When the forming speed of the wet paper web overflows the matching decision range, that is, the forming speed of the wet paper web is too fast, the process of uniform distribution of the slurry by the forming mesh cloth equipment will appear functional decline, the wet paper web fiber arrangement is uneven or has excessive tendency, and then the pressed paper web is prone to uneven stress during drying, resulting in differences in the shrinkage of the paper during the drying process, which ultimately affects the flatness and surface smoothness of the paper. Therefore, by detecting the uniform distribution effect of the wet paper web slurry through machine vision, the rationality of the forming speed of the wet paper web can be controlled in advance, thereby ensuring the flatness of the paper in the material detection of toilet paper. Summary of the invention

[0003] The purpose of the present invention is to provide a method for detecting the forming balance of a wet paper web material of toilet paper so as to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0004] In order to achieve the above object, according to one aspect of the present invention, a method for detecting the forming balance of a wet paper web material of toilet paper is provided, the method comprising the following steps:

[0005] A CCD camera is arranged on the wet paper web conveyor belt; real-time data acquisition is performed through the CCD camera to obtain a paper web image, and the paper web image is preprocessed; the number of uniform failure points and the estimated range of failure points are counted from the paper web image; speed matching decision analysis is performed based on the number of failure points and the estimated range of failure points to form a decision calculation value; and the decision value is used to judge the uniformity of the current wet paper web.

[0006] Furthermore, the method of arranging a CCD camera on the wet paper web conveyor belt is: making the imaging direction of the CCD camera perpendicular to the plane of the wet paper web conveyor belt, using an annular or surface light source as the light source, using a line scanning CCD camera, or replacing the CCD camera with a CMOS camera.

[0007] Furthermore, a paper web image is obtained by real-time data acquisition through a CCD camera, and a method for preprocessing the paper web image is as follows: a CCD camera is used to capture an image of the wet paper web on a wet paper web conveyor belt, the obtained image is used as the paper web image, and the paper web image is grayed; a Gaussian filter is used to process the paper web image to reduce high-frequency noise, and a median filter is used to remove isolated noise points; and the contrast of the image is enhanced through histogram equalization.

[0008] Furthermore, a method for counting the number of uniform failure points and the estimated range of failure points from a paper web image is as follows: by presetting a detection model in the VisionPro program, a number of areas are identified from the paper web image using the detection model and recorded as uniform failure points, the number of which is recorded as the number of uniform failure points, the diameter of the circumscribed circle of the uniform failure points is the sub-estimated range, and the average value of each sub-estimated range is used as the estimated range of failure points.

[0009] Furthermore, the method of forming the decision calculation value by speed matching decision analysis based on the number of failure points and the estimated range of failure points is as follows: set a time period as the detection period R_LMT, R_LMT∈[0.5,1.5] hours, and perform detection at equal intervals within the current R_LMT period; each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i ； The average value and standard deviation of the estimated range of all failure points in the current detection period are recorded as Ave_dis and Std_dis respectively, and the failure range boundary is set to Los_lin = Ave_dis - 3 × Std_dis; if the estimated range of failure points Los_dis obtained at any time i If the time is less than the failure range boundary, the time is defined as a range defect point; the period between any range defect point and the first range defect point in the reverse time direction is defined as a sub-defect interval;

[0010] In the sub-defect interval, the uniform number of failure points is used as the classification variable, and the boxplot function is used to search for outliers in the failure point estimation range under each classification variable, and the sequence corresponding to each outlier is recorded as the outlier sequence Cov.ls; in the outlier sequence, the frequency corresponding to each classification variable is defined as the overflow risk Spris; all the first outlier elements under the classification variable corresponding to the maximum overflow risk are traversed, and the minimum value in the failure point estimation range corresponding to each traversed element is defined as the failure level; for any element in the outlier sequence, if the failure point estimation range at the corresponding moment is less than the failure level, the failure level is updated to the failure point estimation range at that moment, and the average number of uniform failure points and the average failure point estimation range at each detection moment are recorded as Ave_num and Ave_fdi respectively, and the failure level at any detection moment T is calculated. i The corresponding concordant product Codif i =1 / n×(Los_num i -Ave_num)(Los_dis i -Ave_fdi);

[0011] According to each detection time T in any outlier sequence i , the decision calculation value Deval is calculated:

[0012]

[0013] Among them, Codif j It means that within the outlier sequence, the test time T i The moment before T i-1 and the next moment T i+1 The corresponding concordant product, mean is the average function, and ∏ is the product operator symbol.

[0014] The above-mentioned decision calculation value is calculated by accurately identifying the number of uniform failure points, and the frequency of different classification variables locates the critical moment when the quality of the wet paper web has problems, so the quality fluctuations of the wet paper web occurring in real time can be effectively quantified. However, due to the use of box plots to identify outliers, the number of failure points is prone to classification granularity being too coarse during the acquisition process, resulting in the neglect of some subtle quality differences. In particular, when the frequency of the number of failure points needs to be counted to measure the overflow risk, the problem of identifying quality differences is particularly obvious. In order to solve this problem of identifying quality differences and improve the accuracy of the decision calculation value calculation, the present invention proposes a more preferred solution.

[0015] Preferably, the method of forming a decision calculation value by performing speed matching decision analysis through the number of failure points and the estimated range of failure points is: assuming a time period as a detection period R_LMT, R_LMT∈[0.5,1.5] hours, and the values ​​of each uniform number of failure points are respectively used as classification variables;

[0016] Continuously perform detection within the current R_LMT period, and each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i , define the frequency corresponding to each graded variable in the current R_LMT period as the overflow probability,

[0017] The ratio of the overflow probability and the estimated range of the failure point at each detection moment is defined as the operation density gradient Den_grad. Within the detection period R_LMT, the original time series set Orset is composed of a pair of detection moments and operation density gradients. use The plrep and splev functions perform cubic spline interpolation on the original time series set to obtain a smooth curve. The detection moments corresponding to the maximum and minimum values ​​in the obtained curve are risk moments. The risk moment and its corresponding running density gradient form a binary group consisting of a reconstructed time series set Reset. The diff function is used to perform high-order difference operations on the reconstructed time series set of non-stationary time series to obtain a corrected time series set of stationary time series.

[0018] Due to factors such as model accuracy and environmental changes, irregular fluctuations and noise may appear in the data collection results. These fluctuations can be smoothed by cubic spline interpolation, thereby more accurately reflecting the actual change trend of wet paper web quality. In addition, at the moment of significant changes, including sudden failure or quality degradation, the smooth curve generated by cubic spline interpolation can help clearly identify these change points, namely the maximum and minimum values, so as to help locate the critical moment when the wet paper web quality has problems. After high-order difference processing, the autocorrelation between the data points in the corrected time series set is usually significantly reduced. This is because the difference operation removes the correlation components in the sequence to a certain extent. Sequences with high autocorrelation often contain some long-term trends. These trends may not be a true reflection of the change in wet paper web quality, but an inertial phenomenon in the model or preparation process. By reducing the autocorrelation and removing the trend component, we can better focus on the quality fluctuations that occur in real time.

[0019] For any risk moment, the absolute value of the difference between the running density gradients corresponding to the reconstructed time series set and the corrected time series set is defined as the reconstruction error Rerro. All reconstruction errors constitute the first overflow sequence. The risk moments corresponding to the maximum values ​​in the first overflow sequence are taken as overflow points. The set of detection moments between any overflow point and the first overflow point in the reverse time direction is recorded as the peak interval.

[0020] In any peak interval, the reconstruction error of any non-risk moment is calculated by linear interpolation. The median of the reconstruction errors of all moments is recorded as the median error. The difference between the reconstruction error of each moment and the median error is the median deviation. When the median deviation corresponding to any moment is less than half of the difference in the reconstruction errors of the two overflow points in the peak interval to which it belongs, the moment is removed from the peak interval. The set of screened moments is recorded as the correction point set of the peak interval. The data retention rate of the correction point set is defined as the ratio between the number of moments that are not removed and the number of original moments. The absolute value of the median deviation is used to replace the reconstruction errors corresponding to all moments between the two overflow points in the correction point set. The estimated range of failure points corresponding to any moment in the correction point set constitutes the second overflow sequence Scout.

[0021] For any detection time ReT in any correction point set i , calculate the sub-decision calculation value m_Deval:

[0022]

[0023] Where FsT and ScT are the detection time ReT i The starting overflow point and the ending overflow point in the correction point set, hs<> is the harmonic mean function.

[0024] Define the maximum value of all sub-decision calculation values ​​in each correction point set as the decision calculation representative value S_Deval, calculate the weighted average of each decision calculation representative value in the peak point interval as the decision calculation value Deval, and the weight in the weighted average calculation is the data retention rate.

[0025] Furthermore, the method of using the decision value to judge the uniformity of the current wet paper web is: subtract the decision calculation value at the current moment from the first maximum value searched in the reverse time direction, and the difference is recorded as the calculation difference. If the calculation difference at the current moment is larger than the calculation difference at the previous moment and is a positive number, then the current moment is defined to meet the first decision obstacle condition; the preset time period is used as the recent decision segment SR_LMT, and its value range is SR_LMT∈[5,10] minutes; the SR_LMT time period in the reverse time direction at the current moment is defined as the current recent decision segment, and the proportion of moments with positive calculation difference values ​​in the current recent decision segment is the positive calculation proportion; if the positive calculation proportion corresponding to the current moment is greater than the upper quartile value of the positive calculation proportion at each moment in the time period SR_LMT, then the current moment is defined to meet the second decision obstacle condition; if the current moment meets both the first decision obstacle condition and the second decision obstacle condition, then it is determined that the corresponding wet paper web at the current moment does not meet the uniformity requirement.

[0026] Preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

[0027] The present invention also provides a forming balance detection system for a wet paper web material of toilet paper, the forming balance detection system for a wet paper web material of toilet paper comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implementing the steps in the forming balance detection method for a wet paper web material of toilet paper when executing the computer program, the forming balance detection system for a wet paper web material of toilet paper can be run in computing devices such as desktop computers, laptop computers, PDAs, and cloud data centers, the executable system may include, but is not limited to, a processor, a memory, and a server cluster, the processor executing the computer program runs in the following system units:

[0028] An image acquisition setting unit, used for arranging a CCD camera on a wet paper web conveying belt;

[0029] A preprocessing unit, used for acquiring a paper web image by real-time data acquisition through a CCD camera, and preprocessing the paper web image;

[0030] An image analysis unit, used for counting the number of uniform failure points and the estimated range of failure points from the paper web image;

[0031] A decision analysis unit, used to perform speed matching decision analysis based on the number of failure points and the estimated range of failure points to form a decision calculation value;

[0032] The judgment and screening unit is used to judge the uniformity of the current wet paper web by using the decision value.

[0033] The beneficial effects of the present invention are as follows: the present invention provides a method for detecting the forming balance of wet paper web materials of toilet paper, and performs real-time detection and analysis of wet paper web images through machine vision technology, quantifies the risk of functional decline in the process of uniform distribution of pulp by the forming mesh cloth device when the forming speed of the wet paper web is too fast during the production process, and then timely discovers the wet paper web with abnormal flatness and surface smoothness quality of the paper caused by uneven stress, and realizes the accurate evaluation of the distribution uniformity of toilet paper materials during the production process. It not only significantly reduces the loss of raw materials and production costs, and discovers problems from the source of the process flow, but also reduces labor costs by reducing the reliance on manual inspection, effectively avoids the problems of false detection and missed detection caused by human factors, and significantly improves production efficiency, providing an efficient and reliable detection scheme for flatness and smoothness quality assurance for the toilet papermaking industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0035] Figure 1 Shown is a flow chart of a method for detecting the forming balance of a wet paper web material of toilet paper;

[0036] Figure 2 Shown is a structural diagram of a forming balance detection system for a wet paper web material of toilet paper. DETAILED DESCRIPTION

[0037] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0038] like Figure 1 The flowchart of the method for detecting the forming balance of a wet paper web material of toilet paper is shown below. Figure 1 A method for detecting the forming balance of a wet paper web material of toilet paper according to an embodiment of the present invention is described below. The method comprises the following steps:

[0039] A CCD camera is arranged on the wet paper web conveyor belt; real-time data acquisition is performed through the CCD camera to obtain a paper web image, and the paper web image is preprocessed; the number of uniform failure points and the estimated range of failure points are counted from the paper web image; speed matching decision analysis is performed based on the number of failure points and the estimated range of failure points to form a decision calculation value; and the decision value is used to judge the uniformity of the current wet paper web.

[0040] Furthermore, the method of arranging a CCD camera on the wet paper web conveyor belt is: making the imaging direction of the CCD camera perpendicular to the plane of the wet paper web conveyor belt, using an annular or surface light source as the light source, using a line scanning CCD camera, or replacing the CCD camera with a CMOS camera.

[0041] Furthermore, a paper web image is obtained by real-time data acquisition through a CCD camera, and a method for preprocessing the paper web image is as follows: a CCD camera is used to capture an image of the wet paper web on a wet paper web conveyor belt, the obtained image is used as the paper web image, and the paper web image is grayed; a Gaussian filter is used to process the paper web image to reduce high-frequency noise, and a median filter is used to remove isolated noise points; and the contrast of the image is enhanced through histogram equalization to make the edge of the paper web more obvious.

[0042] Furthermore, a method for counting the number of uniform failure points and the estimated range of failure points from a paper web image is as follows: by presetting a detection model in the VisionPro program, a number of areas are identified from the paper web image using the detection model and recorded as uniform failure points, the number of which is recorded as the number of uniform failure points, the diameter of the circumscribed circle of the uniform failure points is the sub-estimated range, and the average value of each sub-estimated range is used as the estimated range of failure points.

[0043] In the process of obtaining uniform failure points, the paper image is binarized by the Ostu threshold method, connected component marking is performed, and several foreground areas are identified as uniform failure points; connected component marking means that in a binary image, continuous areas composed of pixels with the same value of 255 are marked as areas with the same serial number.

[0044] Furthermore, the method of forming a decision calculation value by speed matching decision analysis based on the number of failure points and the estimated range of failure points is as follows: set a time period as the detection period R_LMT, R_LMT∈[0.5,1.5] hours, perform detection at equal intervals within the current R_LMT period, that is, obtain the number of failure points and the estimated range of failure points once, and record the detection event scale as the detection time T i ; T i There is a one-to-one correspondence with i; each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i, where the binary value refers to the element in the binary group; the average value and standard deviation of the estimated range of all failure points in the current detection period are recorded as Ave_dis and Std_dis respectively, and the failure range boundary is set to Los_lin = Ave_dis-3×Std_dis; if the estimated range of failure points Los_dis obtained at any time i If the time is less than the failure range boundary, the time is defined as a range defect point; the period between any range defect point and the first range defect point in the reverse time direction is defined as a sub-defect interval;

[0045] In the sub-defect interval, the uniform number of failure points is used as the classification variable, and the boxplot function is used to search for outliers in the failure point estimation range under each classification variable, and the sequence of each outlier corresponding to the moment is recorded as the outlier sequence Cov.ls; in the outlier sequence, the frequency corresponding to each classification variable is defined as the overflow risk Spris, where the frequency corresponding to each uniform number of failure points refers to the ratio of the number of outliers in the same classification variable to the number of elements in the outlier sequence. ; Traverse all the first outlier elements under the classification variable corresponding to the maximum overflow risk, and define the minimum value in the failure point estimation range corresponding to each traversed element as the failure level; the first outlier element refers to the element in the outlier sequence; for any element in the outlier sequence, if the failure point estimation range at the corresponding moment is less than the failure level, then the failure level is updated to the failure point estimation range at that moment, and the average values ​​of the uniform number of failure points and the failure point estimation range at each detection moment are recorded as Ave_num and Ave_fdi respectively, and the average value of the uniform number of failure points and the failure point estimation range at any detection moment T is calculated. i The corresponding concordant product Codif i =1 / n×(Los_num i -Ave_num)(Los_dis i -Ave_fdi);

[0046] According to each detection time T in any outlier sequence i , the decision calculation value Deval is calculated:

[0047]

[0048] Among them, Codif j It means that within the outlier sequence, the test time T i The moment before T i-1 and the next moment T i+1 The corresponding concordant product, if T i is the initial detection time or the end detection time, and the denominator is defined as (codif j ―codif i ) 2,The initial detection time or the end detection time refers to the first and last detection time in the R_LMT period, exp() is an exponential function with the natural constant e as the base, and the boxplot function is called by the matplotlib library in python;

[0049] Preferably, the method of forming a decision calculation value by performing speed matching decision analysis through the number of failure points and the estimated range of failure points is: assuming a time period as a detection period R_LMT, R_LMT∈[0.5,1.5] hours, and the values ​​of each uniform number of failure points are respectively used as classification variables;

[0050] Continuously perform detection within the current R_LMT period, and each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i ,Define the frequency corresponding to each grading variable in the current R_LMT period as the overflow probability, where the frequency corresponding to any grading variable refers to the proportion of the number of occurrences of a grading variable to the total number of detections in the detection period;

[0051] Define the ratio of the underflow probability and the estimated range of the failure point at each detection moment as the operation density gradient Den_grad. Within the detection period R_LMT, the original time series set Orset is composed of a binary group consisting of the detection moment and the operation density gradient. The original time series set is interpolated with cubic spline using the plrep and splev functions to obtain a smooth curve. The detection moments corresponding to the maximum and minimum values ​​in the obtained curve are risk moments. When the moments corresponding to the maximum and minimum values ​​are not detection moments, the previous detection moment and the next detection moment with the shortest interval to the moment are defined as risk moments. The binary group consisting of the risk moment and its corresponding operation density gradient constitutes the reconstructed time series set Reset. Use the diff function to perform high-order difference operations on the reconstructed time series set of non-stationary time series to obtain the corrected time series set of stationary time series.

[0052] For any risk moment, the absolute value of the difference between the running density gradients corresponding to the reconstructed time series set and the corrected time series set is defined as the reconstruction error Rerro. All reconstruction errors constitute the first overflow sequence. The risk moments corresponding to the maximum values ​​in the first overflow sequence are taken as overflow points. The set of detection moments between any overflow point and the first overflow point in the reverse time direction is recorded as the peak interval.

[0053] In any peak point interval, the reconstruction error of any non-risk moment is calculated by linear interpolation. The median of the reconstruction errors of all moments is recorded as the median error. The difference between the reconstruction error of each moment and the median error is the median deviation. When the median deviation corresponding to any moment is less than half of the difference in the reconstruction errors of the two overflow points in the peak point interval to which it belongs, the moment is removed from the peak point interval. The set of screened moments is recorded as the correction point set of the peak point interval. The data retention rate of the correction point set is defined as the ratio between the number of moments that have not been removed and the number of original moments. The absolute value of the median deviation is used to replace the reconstruction errors corresponding to all moments between the two overflow points in the correction point set. The reconstruction errors corresponding to the two overflow points are not rewritten. The estimated range of failure points corresponding to any moment in the correction point set constitutes the second overflow sequence Scout.

[0054] For any detection time ReT in any correction point set i , calculate the sub-decision calculation value m_Deval:

[0055]

[0056] Where FsT and ScT are the detection time ReT i The starting overflow point and the ending overflow point in the correction point set, hs<> is the harmonic mean function, and exp() is the exponential function with the natural constant e as the base;

[0057] Define the maximum value of all sub-decision calculation values ​​in each correction point set as the decision calculation representative value S_Deval, calculate the weighted average of each decision calculation representative value in the peak point interval as the decision calculation value Deval, and the weight in the weighted average calculation is the data retention rate.

[0058] The decision calculation value is calculated as follows:

[0059]

[0060] Where Num is the number of correction point sets in the current detection period, and the weight used is the data retention rate of each correction point set. j , Darat j is the decision calculation value and data retention rate of the jth correction point set.

[0061] The splrep and splev functions in Python are called by the matplotlib library, and the diff function is called by the statsmodels library.

[0062] Furthermore, the method of using the decision value to judge the uniformity of the current wet paper web is: subtract the decision calculation value at the current moment from the first maximum value searched in the reverse time direction, and the difference is recorded as the calculation difference. If the calculation difference at the current moment is larger than the calculation difference at the previous moment and is a positive number, then the current moment is defined to meet the first decision obstacle condition; the preset time period is used as the recent decision segment SR_LMT, and its value range is SR_LMT∈[5,10] minutes; the SR_LMT time period in the reverse time direction at the current moment is defined as the current recent decision segment, and the proportion of moments with positive calculation difference values ​​in the current recent decision segment is the positive calculation proportion; if the positive calculation proportion corresponding to the current moment is greater than the upper quartile value of the positive calculation proportion at each moment in the time period SR_LMT, then the current moment is defined to meet the second decision obstacle condition; if the current moment meets both the first decision obstacle condition and the second decision obstacle condition, then it is determined that the corresponding wet paper web at the current moment does not meet the uniformity requirement.

[0063] An embodiment of the present invention provides a forming balance detection system for a wet paper web material of toilet paper, such as Figure 2 Shown is a structural diagram of a forming balance detection system for a wet paper web material of toilet paper of the present invention. The forming balance detection system for a wet paper web material of toilet paper of this embodiment comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned forming balance detection method embodiment for a wet paper web material of toilet paper are implemented.

[0064] The system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to run in the following units of the system:

[0065] An image acquisition setting unit, used for arranging a CCD camera on a wet paper web conveying belt;

[0066] A preprocessing unit, used for acquiring a paper web image by real-time data acquisition through a CCD camera, and preprocessing the paper web image;

[0067] An image analysis unit, used for counting the number of uniform failure points and the estimated range of failure points from the paper web image;

[0068] A decision analysis unit, used to perform speed matching decision analysis based on the number of failure points and the estimated range of failure points to form a decision calculation value;

[0069] The judgment and screening unit is used to judge the uniformity of the current wet paper web by using the decision value.

[0070] The forming balance detection system of the wet paper web material of toilet paper can be run in computing devices such as desktop computers, laptop computers, PDAs and cloud servers. The forming balance detection system of the wet paper web material of toilet paper can be operated in a system that may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the example is only an example of a forming balance detection system of a wet paper web material of toilet paper, and does not constitute a limitation on the forming balance detection system of a wet paper web material of toilet paper. It can include more or fewer components than the example, or combine certain components, or different components. For example, the forming balance detection system of the wet paper web material of toilet paper can also include input and output devices, network access devices, buses, etc.

[0071] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the operating system of the forming balance detection system for the wet paper web material of toilet paper, and uses various interfaces and lines to connect various parts of the operating system of the forming balance detection system for the wet paper web material of toilet paper.

[0072] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the forming balance detection system of the wet paper web material of toilet paper by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0073] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. A method for detecting the forming balance of a wet paper web material of toilet paper, characterized in that: The method comprises the following steps: arranging a CCD camera on a wet paper web conveyor belt; acquiring a paper web image by real-time data acquisition through the CCD camera, and preprocessing the paper web image; counting the number of uniform failure points and the estimated range of failure points from the paper web image; performing speed matching decision analysis based on the number of failure points and the estimated range of failure points to form a decision calculation value; and judging the uniformity of the current wet paper web using the decision value; The method of forming a decision calculation value by performing speed matching decision analysis based on the number of failure points and the estimated range of failure points is as follows: defining the failure range boundary according to the estimated range of failure points within a preset time period, searching and locating the range defect points using the failure range boundary, constructing a sub-defect interval through the range defect points, using the uniform number of failure points as a classification variable, constructing an outlier sequence according to the outlier values ​​of the estimated range of failure points under different classification variables in the sub-defect interval, and calculating the decision calculation value using the outlier sequence.

2. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that: The method of arranging a CCD camera on a wet paper web conveyor belt is as follows: the imaging direction of the CCD camera is perpendicular to the plane of the wet paper web conveyor belt, the light source is an annular or surface light source, the CCD camera is a line scan CCD camera, or a CMOS camera is used instead of the CCD camera.

3. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that: The paper web image is obtained by real-time data acquisition through a CCD camera, and the method for preprocessing the paper web image is as follows: the wet paper web on the wet paper web conveyor belt is imaged by a CCD camera, the obtained image is used as the paper web image, and the paper web image is grayed; the paper web image is processed by a Gaussian filter to reduce high-frequency noise, and a median filter is used to remove isolated noise points; and the image contrast is enhanced by histogram equalization.

4. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that ,The method of counting the number of uniform failure points and the estimated range of ,failure points from the paper web image is as follows: by presetting the detection model in the VisionPro program, ,using the detection model to identify several areas from the paper web image and ,record them as uniform failure points, and their number is recorded as the number of uniform ,failure points, the circumscribed circle diameter of the uniform failure points is the sub-estimated range, and the average value of each sub-estimated range is the estimated range of ,failure points.

5. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that: The method of forming the decision calculation value by speed matching decision analysis based on the number of failure points and the estimated range of failure points is as follows: suppose a time period is used as the detection period R_LMT, R_LMT∈[0.5,1.5] hours, and detection is performed at equal intervals within the current R_LMT period. At each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i , record the average value and standard deviation of all failure point prediction ranges in the current detection period as Ave_dis and Std_dis respectively, and set the failure range boundary as Los_lin = Ave_dis - 3 × Std_dis; if the failure point prediction range Los_dis obtained at any time i If the time is less than the failure range boundary, the time is defined as a range defect point; the period between any range defect point and the first range defect point in the reverse time direction is defined as a sub-defect interval; In the sub-defect interval, the uniform number of failure points is used as the classification variable, and the boxplot function is used to search for outliers in the failure point estimation range under each classification variable, and the sequence corresponding to each outlier is recorded as the outlier sequence Cov.ls; in the outlier sequence, the frequency corresponding to each classification variable is defined as the overflow risk Spris; all the first outlier elements under the classification variable corresponding to the maximum overflow risk are traversed, and the minimum value in the failure point estimation range corresponding to each traversed element is defined as the failure level; for any element in the outlier sequence, if the failure point estimation range at the corresponding moment is less than the failure level, the failure level is updated to the failure point estimation range at that moment, and the average number of uniform failure points and the average failure point estimation range at each detection moment are recorded as Ave_num and Ave_fdi respectively, and the failure level at any detection moment T is calculated. i The corresponding synergistic product is used to calculate the decision calculation value Deval based on the overflow risk and the synergistic product.

6. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that: The method of forming a decision calculation value by speed matching decision analysis through the number of failure points and the estimated range of failure points can be replaced by: assuming a time period as the detection period R_LMT, R_LMT∈[0.5,1.5] hours, and the values ​​of each uniform number of failure points are respectively used as classification variables; Continuously perform detection within the current R_LMT period, and each detection time T i The corresponding detection binary values ​​are the uniform failure point number Los_num i And the failure point estimation range Los_dis i ,Define the frequency corresponding to each graded variable in the current R_LMT period as the overflow probability; Define the ratio of the underflow probability and the estimated range of the failure point at each detection moment as the operating density gradient Den_grad. In the detection period R_LMT, the binary group consisting of the detection moment and the operating density gradient constitutes the original time series set Orset. Use the plrep and splev functions to perform cubic spline interpolation on the original time series set to obtain a smooth curve. The detection moments corresponding to the maximum and minimum values ​​in the obtained curve are risk moments. The binary group consisting of the risk moment and its corresponding operating density gradient constitutes the reconstructed time series set Reset. Use the diff function to perform high-order difference operations on the reconstructed time series set of non-stationary time series to obtain the corrected time series set of stationary time series. For any risk moment, the absolute value of the difference between the running density gradients corresponding to the reconstructed time series set and the corrected time series set is defined as the reconstruction error Rerro. All reconstruction errors constitute the first overflow sequence. The risk moments corresponding to the maximum values ​​in the first overflow sequence are taken as overflow points. The set of detection moments between any overflow point and the first overflow point in the reverse time direction is recorded as the peak interval. In any peak interval, the reconstruction error of any non-risk moment is calculated by linear interpolation. The median of the reconstruction errors of all moments is recorded as the median error. The difference between the reconstruction error of each moment and the median error is the median deviation. When the median deviation corresponding to any moment is less than half of the difference in the reconstruction errors of the two overflow points in the peak interval to which it belongs, the moment is removed from the peak interval. The set of screened moments is recorded as the correction point set of the peak interval. The data retention rate of the correction point set is defined as the ratio between the number of moments that are not removed and the number of original moments. The absolute value of the median deviation is used to replace the reconstruction errors corresponding to all moments between the two overflow points in the correction point set. The estimated range of failure points corresponding to any moment in the correction point set constitutes the second overflow sequence Scout. For any detection time ReT in any correction point set i , the sub-decision calculation value is calculated according to the second overflow sequence and the reconstruction error; the maximum value of all sub-decision calculation values ​​in each correction point set is defined as the decision calculation representative value S_Deval, and the weighted average of each decision calculation representative value in the peak point interval is calculated as the decision calculation value Deval, and the weight in the weighted average calculation is the data retention rate.

7. The method for detecting the forming balance of a wet paper web material of toilet paper according to claim 1, characterized in that: The method of using the decision value to judge the uniformity of the current wet paper web is as follows: the decision calculation value at the current moment is subtracted from the first maximum value searched in the reverse time direction, and the difference is recorded as the calculation difference. If the calculation difference at the current moment is larger than the calculation difference at the previous moment and is a positive number, the current moment is defined to meet the first decision obstacle condition; a time period is preset as the recent decision segment SR_LMT; the SR_LMT time period in the reverse time direction at the current moment is defined as the current recent decision segment, and the proportion of moments with positive calculation difference values ​​in the current recent decision segment is the positive calculation proportion; if the positive calculation proportion corresponding to the current moment is greater than the upper quartile value of the positive calculation proportion at each moment in the time period SR_LMT, the current moment is defined to meet the second decision obstacle condition; if the current moment meets both the first decision obstacle condition and the second decision obstacle condition, it is determined that the wet paper web corresponding to the current moment does not meet the uniformity requirement.

8. A forming balance detection system for a wet paper web material of toilet paper, characterized in that: The forming balance detection system of the wet paper web material of toilet paper comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the forming balance detection method of the wet paper web material of toilet paper described in any one of claims 1 to 7 are implemented. The forming balance detection system of the wet paper web material of toilet paper runs on a desktop computer, a laptop computer, a PDA, and a computing device in a cloud data center.

Images