On-line detection method and system for tear strength of thermo-sensitive paper based on machine vision

Abstract

The invention discloses an online detection method and system for the tear strength of thermo-sensitive paper based on machine vision, and relates to the technical field of thermo-sensitive paper tear strength detection.The method comprises the following steps that a paper image is collected; recording an audio waveform generated when the thermo-sensitive paper is flapped; extracting surface features of the thermo-sensitive paper in the paper image; acquiring sound wave characteristics of the thermo-sensitive paper; the method comprises the following steps: constructing a thermo-sensitive paper tear strength on-line detection model, collecting surface features and sound wave features of thermo-sensitive paper with known quality, and carrying out deep learning; analyzing the surface characteristics and the sound wave characteristics through a thermo-sensitive paper tear strength online detection model, and judging whether the tear strength of the thermo-sensitive paper is qualified or not; the invention is used for solving the problems that the productivity of the thermo-sensitive paper is reduced and the yield of the thermo-sensitive paper cannot be guaranteed due to the fact that the thermo-sensitive paper needs to be damaged to detect the tearing strength and a sampling mode is adopted for detection in the existing thermo-sensitive paper tearing strength detection technology.

Description

Technical Field

[0001] This invention relates to the field of thermal paper tear resistance testing technology, specifically to a machine vision-based online method and system for thermal paper tear resistance testing. Background Technology

[0002] Thermal paper tear resistance testing technology refers to a set of methods and means used to quantitatively measure the ability of thermal paper to resist tearing and propagation by external forces. It usually requires applying a controllable force and measuring the force required for the paper to continue to tear, thereby objectively assessing its mechanical durability and physical quality.

[0003] Existing thermal paper tear strength testing technologies typically require removing the thermal paper from the production line after production and tearing it with a tensile testing machine to assess its tear strength. This method can only evaluate the tear strength of thermal paper through sampling, and it requires damaging the thermal paper, which leads to a decrease in production capacity. Furthermore, sampling only reflects the whole from a part, which has a certain degree of error and cannot ensure the accuracy of the test results. In some cases, it cannot correctly detect unqualified thermal paper, and the yield rate cannot be guaranteed. Existing thermal paper tear strength testing technologies also have the problems of requiring the thermal paper to be damaged for tear strength testing and using sampling methods, which reduces the production capacity of thermal paper and cannot guarantee the yield rate. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. It involves installing a light source and camera on a thermal paper production line to capture paper images, then installing a paper-beating device on the production line to beat the thermal paper and record the waveform of the sound generated when the thermal paper is beaten (named audio waveform). Feature analysis is then performed on the paper images to extract the surface features of the thermal paper, and feature analysis is also performed on the audio waveform to obtain the acoustic wave features of the thermal paper. An online tear resistance detection model for thermal paper is then constructed. Surface features of thermal paper of known quality are collected and deep learning is performed to obtain surface reference features. Simultaneously, acoustic wave features of thermal paper of known quality are collected and deep learning is performed to obtain acoustic wave reference features. Finally, the online tear resistance detection model analyzes the surface features and acoustic wave features to determine whether the tear resistance of the thermal paper is qualified. This addresses the problem that existing thermal paper tear resistance detection technologies require damaging the thermal paper for tear resistance testing and use sampling methods, which reduce thermal paper production capacity and compromise the yield rate.

[0005] To achieve the above objectives, in a first aspect, this application provides an online method for detecting the tear resistance of thermal paper based on machine vision, comprising the following steps: Light sources and cameras are installed on the thermal paper production line to capture images of the paper. A paper tapping device is installed on the thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped, which is named audio waveform. Feature analysis is performed on paper images to extract the surface features of thermal paper from the paper images; Feature analysis of audio waveforms is performed to obtain the sound wave characteristics of thermal paper; An online detection model for the tear resistance of thermal paper was constructed, and the surface and acoustic features of thermal paper of known quality were collected and deep learning was performed. The surface and acoustic characteristics of thermal paper are analyzed using an online tear resistance testing model to determine whether the tear resistance of the thermal paper is up to standard.

[0006] Furthermore, installing a light source and a camera on the thermal paper production line to capture images of the paper includes the following sub-steps: The light source is a flicker-free LED linear light source, and the camera is a line scan camera; The LED linear light source illuminates the thermal paper at a first angle, and the linear array camera is positioned directly above the thermal paper to capture an image of the paper.

[0007] Furthermore, installing a paper tapping device on the thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped includes the following sub-steps: The paper tapping device includes a front conveyor belt, a rear conveyor belt, a front fixing device, a rear fixing device, and a tapping device. The front fixing device is installed above the front conveyor belt, and the rear fixing device is installed above the rear conveyor belt. Before the thermal paper is tapped, the front conveyor belt stops running, the front fixing device presses down to fix the front end of the thermal paper, the rear conveyor belt stops running, the rear fixing device presses down to fix the rear end of the thermal paper, and then the rear conveyor belt is started. The rear fixing device can move synchronously with the rear conveyor belt. At this time, a protrusion will be generated on the thermal paper between the front fixing device and the rear fixing device. The tapping device taps the protrusion and collects the waveform of the sound generated by the thermal paper during the tapping to obtain the audio waveform.

[0008] Further, feature analysis is performed on the paper image to extract the surface features of the thermal paper, including the following sub-steps: The paper image is converted to grayscale to obtain a grayscale image; Obtain the gray-level co-occurrence matrix of the gray-level image, and extract the contrast and correlation from the gray-level co-occurrence matrix as surface features of the thermal paper.

[0009] Further, feature analysis of the audio waveform is performed to obtain the acoustic wave characteristics of the thermal paper, including the following sub-steps: The audio waveform is stored in an audio graph, where the X-axis represents time and the Y-axis represents amplitude. The coordinates of the peaks of the audio waveform are obtained and named peak points, and the coordinates of the troughs of the audio waveform are obtained and named trough points. The peaks and troughs are numbered in chronological order, using the symbol F. n and G n This represents the expression, where n is a non-zero natural number and n is the index of F and G. Typically, the number of peaks and valleys is the same, and G... n For located at F n The first valley after that; F through the straight line n With G n Connect the lines and mark the resulting auxiliary lines as L. n , obtain L n The slope, denoted as K n ; Through a straight line to two adjacent F n Connecting the segments yields a single line segment, which is named the peak analysis line. F n The corresponding time stamp is T n , will K n With T n Related, with T n K is the horizontal axis. n Establish a Cartesian coordinate system for the vertical axis, named the Peak-Valley Analysis Chart, and set K... n According to T n Enter the peak-valley analysis chart and name the coordinate points in the chart as peak-valley analysis points; K n With T n The coordinate point formed is labeled P. n Through a straight line to two adjacent P n Connecting the segments yields a broken line segment, which is named the Peak-Valley Analysis Broken Line. The peak analysis line and the peak-valley analysis line are the characteristics of the sound wave.

[0010] Furthermore, an online detection model for the tear resistance of thermal paper is constructed. This involves collecting surface and acoustic features of thermal paper of known quality and performing deep learning, including the following sub-steps: An online detection model for the tear resistance of thermal paper was constructed. The surface features of thermal paper of known quality were collected and deep learning was performed to obtain surface reference features. Acoustic wave characteristics of thermal paper of known quality are collected and deep learning is performed to obtain acoustic wave reference characteristics.

[0011] Furthermore, an online detection model for the tear resistance of thermal paper is constructed. Surface features of thermal paper of known quality are collected and deep learning is performed to obtain surface reference features, including the following sub-steps: An online detection model for the tear resistance of thermal paper is constructed. Thermal paper of known quality is used as sample paper. The quality of the sample paper is named as sample quality. The surface features of the sample paper are extracted and named as surface sample features. The sample quality corresponds to the surface sample features. The contrast and correlation in the surface sample features are named sample contrast and sample correlation, respectively. The range of sample contrast with acceptable sample quality is used to obtain a first reference range, and the range of sample correlation with acceptable sample quality is used to obtain a second reference range. The first reference range and the second reference range together constitute the surface reference feature.

[0012] Furthermore, the acoustic wave characteristics of thermal paper of known quality are collected and deep learning is performed to obtain acoustic wave reference characteristics, including the following sub-steps: Acoustic characteristics of sample paper with qualified sample quality were collected and named as acoustic sample characteristics. The peak analysis line and the peak-valley analysis line in the acoustic sample characteristics were named as peak sample line and peak-valley sample line, respectively. Name the leftmost endpoint of the peak sample line as the peak start point, name the leftmost endpoint of the valley sample line as the valley start point, and name the turning points and endpoints in the peak and valley sample lines as line vertices. Overlap the peak start points of all peak sample lines and name the overlapping image a peak learning graph; overlap the peak and valley start points of all valley sample lines and name the overlapping image a valley learning graph. When performing deep learning on peak learning graphs or valley learning graphs, the peak learning graph or valley learning graph being analyzed is named the target analysis graph, and the X-axis and Y-axis corresponding to the target analysis graph are named the first axis and the second axis, respectively. For each value of the first axis, obtain the topmost and bottommost polyline vertices and name them the upper limit point and lower limit point, respectively. The upper and lower limits are numbered from left to right, and are respectively represented by the symbol U. i and D jThis indicates that i and j are both non-zero natural numbers, i is the index of U, and j is the index of D; Set an auxiliary number h, which is initially 1. Starting with i=1, pass through the straight line U i with U i+h Connect the points to obtain the upper limit analysis line. Determine whether the upper limit analysis line intersects with any part of the target analysis diagram except for the upper limit point. If yes, output an upper limit error signal; otherwise, output an upper limit correct signal. If an upper limit error signal is output, the upper limit analysis line will be deleted, and U will be... i+h Delete, increment h and reanalyze the upper limit analysis line, repeat the judgment until the upper limit is correctly output; if the upper limit is correctly output, retain the upper limit analysis line, and increment i by 1, then judge U. i Does it exist, if U i If it does not exist, increment i again; if U i If it exists, reset h to 1 and analyze another upper limit analysis line again until the maximum value of i is reached, to obtain different upper limit analysis lines. The upper limit analysis lines form a new polyline segment, which is named the upper limit reference line. Reset h to 1, starting with j=1, and move D through a straight line. j With D j+h Connect the points to obtain the lower limit analysis line. Determine whether the lower limit analysis line intersects with any part of the target analysis diagram except for the lower limit point. If yes, output the lower limit error signal; otherwise, output the lower limit correct signal. If a lower limit error signal is output, the lower limit analysis line will be deleted, and D will be... j+h Delete, increment h and re-analyze the lower limit analysis line, repeat the judgment until the lower limit correct signal is output; if the lower limit correct signal is output, retain the lower limit analysis line, and increment j by 1, then judge D. j Does it exist if D j If it does not exist, increment j again. If D j If it exists, reset h to 1 and analyze another lower limit analysis line again until the maximum value of j is reached, to obtain different lower limit analysis lines. The lower limit analysis lines form a new line segment, which is named the lower limit reference line. Name the leftmost endpoints of the upper and lower reference lines as the upper left endpoint and the lower left endpoint, respectively. Name the rightmost endpoints of the upper and lower reference lines as the upper right endpoint and the lower right endpoint, respectively. Connect the upper left endpoint and the lower left endpoint, and connect the upper right endpoint and the lower right endpoint to obtain a closed region, which is named the target reference feature. The target reference features obtained from the analysis of the peak learning map and the peak-valley learning map are named peak reference features and peak-valley reference features, respectively. The peak reference features and peak-valley reference features together constitute the sound wave reference features.

[0013] Furthermore, by analyzing surface features and acoustic characteristics using an online tear resistance testing model for thermal paper, the determination of whether the tear resistance of the thermal paper is qualified includes the following sub-steps: The surface features and acoustic features to be analyzed are named Real-time Surface Features and Real-time Acoustic Features, respectively. The contrast and correlation in the Real-time Surface Features are named Real-time Contrast and Real-time Correlation, respectively. The peak analysis line and peak-valley analysis line in the Real-time Acoustic Features are named Real-time Peak Line and Real-time Peak-valley Line, respectively. If the real-time contrast is within the first reference range, the real-time correlation is within the second reference range, and all real-time peak lines are within the peak reference features and all real-time trough lines are within the trough reference features, then the tear resistance strength is qualified; otherwise, the tear resistance strength is unqualified.

[0014] Secondly, this application provides an online detection system for the tear resistance of thermal paper based on machine vision, including an image acquisition module, an audio acquisition module, a surface feature extraction module, a sound wave feature extraction module, a feature analysis module, and a tear resistance determination module; the image acquisition module, audio acquisition module, surface feature extraction module, sound wave feature extraction module, and tear resistance determination module are respectively connected to the feature analysis module for data transmission. The image acquisition module is used to install a light source and a camera on the thermal paper production line to acquire images of the paper. The audio acquisition module is used to install a paper tapping device on the thermal paper production line, which taps the thermal paper and records the waveform of the sound generated when the thermal paper is tapped, named audio waveform. The surface feature extraction module is used to perform feature analysis on the paper image and extract the surface features of the thermal paper in the paper image; The sound wave feature extraction module is used to perform feature analysis on the audio waveform to obtain the sound wave features of the thermal paper; The feature analysis module is used to construct an online detection model for the tear resistance of thermal paper, collect the surface features and acoustic features of thermal paper of known quality and perform deep learning; The tear resistance strength judgment module is used to analyze surface features and acoustic features through an online detection model for the tear resistance strength of thermal paper to determine whether the tear resistance strength of the thermal paper is qualified.

[0015] The beneficial effects of this invention are as follows: This invention installs a light source and a camera on a thermal paper production line to acquire paper images. Then, a paper-beating device is installed on the thermal paper production line to beat the thermal paper and record the waveform of the sound produced when the thermal paper is beaten, which is named the audio waveform. Feature analysis is then performed on the paper images to extract the surface features of the thermal paper in the paper images. At the same time, feature analysis is performed on the audio waveform to obtain the sound wave characteristics of the thermal paper. The advantage is that there are microscopic textures on the surface of thermal paper, and the surface features are the manifestation of the microscopic textures. Moreover, the sounds emitted when the paper with different tear resistance is beaten are also different. By obtaining the sound wave characteristics, the accuracy and effectiveness of thermal paper tear resistance detection are improved. This invention constructs an online detection model for the tear resistance of thermal paper, collects surface features of thermal paper of known quality and performs deep learning to obtain surface reference features, and simultaneously collects acoustic features of thermal paper of known quality and performs deep learning to obtain acoustic reference features. Finally, the online detection model for the tear resistance of thermal paper analyzes the surface features and acoustic features to determine whether the tear resistance of the thermal paper is qualified. The advantage is that although the sound differences between papers of the same material are small, computers can still extract the subtle differences. By combining machine vision and voiceprint recognition through deep learning, the accuracy and effectiveness of online detection of the tear resistance of thermal paper are further improved. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system of the present invention; Figure 2 This is a schematic diagram of the paper tapping device of the present invention; Figure 3 This is a schematic diagram of the audio waveform of the present invention; Figure 4 For the L of the present invention n A schematic diagram; Figure 5 This is a schematic diagram of the peak analysis polygon of the present invention; Figure 6 This is a schematic diagram of the peak-valley analysis diagram of the present invention; Figure 7 This is a schematic diagram of the peak learning graph of the present invention; Figure 8 This is a schematic diagram of the upper and lower limits of the present invention; Figure 9 This is a schematic diagram of the peak reference feature of the present invention; Figure 10 This is a flowchart of the steps of the method of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, please refer to Figure 1 As shown, this application provides an online detection system for the tear resistance of thermal paper based on machine vision, including an image acquisition module, an audio acquisition module, a surface feature extraction module, a sound wave feature extraction module, a feature analysis module, and a tear resistance determination module; the image acquisition module, audio acquisition module, surface feature extraction module, sound wave feature extraction module, and tear resistance determination module are respectively connected to the feature analysis module for data transmission. The image acquisition module is used to install a light source and a camera on the thermal paper production line to acquire images of the paper. The image acquisition module is configured with an image acquisition strategy, which includes: The light source uses a flicker-free LED linear light source, and the camera is a line scan camera; An LED linear light source illuminates the thermal paper at a first angle, and a line scan camera positioned directly above the thermal paper captures an image of the paper. In practical applications, the line scan camera uses ultra-high resolution. In this embodiment, an 8K resolution line scan camera is used. The line scan camera only collects data from one row of pixels, and this row contains 8192 pixels. This allows the acquisition of micron-level texture information on the surface of the thermal paper. Under the illumination of the LED linear light source, the texture information exhibits a certain degree of light and dark contrast. The flicker-free LED linear light source and the line scan camera are both existing devices, and will not be described in detail in this embodiment. The first angle is adjusted by the tester to ensure that there is light and dark contrast in the acquired paper image.

[0019] The audio acquisition module is used to install a paper tapping device on the thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped, which is named audio waveform. The audio acquisition module is configured with an audio acquisition strategy, which includes: Please see Figure 2 As shown, the paper tapping device includes a front conveyor belt, a rear conveyor belt, a front fixing device, a rear fixing device, and a tapping device. The front fixing device is installed above the front conveyor belt, and the rear fixing device is installed above the rear conveyor belt. Before the thermal paper is tapped, the front conveyor belt stops running, the front fixing device presses down to fix the front end of the thermal paper, the rear conveyor belt stops running, the rear fixing device presses down to fix the rear end of the thermal paper, and then the rear conveyor belt is started. The rear fixing device can move synchronously with the rear conveyor belt. At this time, a protrusion will be generated on the thermal paper between the front fixing device and the rear fixing device. The tapping device taps the protrusion and collects the waveform of the sound generated by the thermal paper during the tapping to obtain the audio waveform. In practical applications, the front conveyor belt, rear conveyor belt, front fixing device, and rear fixing device in the paper tapping device are as follows: Figure 2 As shown, the tapping device is located between the front and rear fixing devices. The tapping device is a robotic arm, and the axis of rotation of the robotic arm is far from the thermal paper. The long robotic arm taps the thermal paper, ensuring that the axis of rotation of the robotic arm does not affect the recording of the audio waveform. The different tear resistance of the same type of paper when tapped will also cause slight differences in the audio waveform. The difference in tear resistance is caused by the different structures or materials of the paper during production. The audio waveform generated by the paper being tapped is also affected by the material and structure of the paper. Therefore, when the tear resistance of the paper is different, its audio waveform will also produce slight changes under controlled variables.

[0020] The surface feature extraction module is used to perform feature analysis on paper images and extract the surface features of thermal paper from the paper images; The surface feature extraction module is configured with surface feature extraction strategies, which include: The paper image is converted to grayscale to obtain a grayscale image; Obtain the gray-level co-occurrence matrix of the gray-level image, and extract the contrast and correlation in the gray-level co-occurrence matrix as surface features of the thermal paper; In practical applications, the acquisition and construction of gray-level co-occurrence matrix are existing technologies, and will not be described in detail in this embodiment. The analysis of gray-level co-occurrence matrix is ​​based on machine vision methods. However, in the actual analysis process, machine vision still has certain drawbacks. Therefore, this embodiment integrates voiceprint recognition on the basis of machine vision, and performs a deeper analysis of the tear resistance of thermal paper by extracting subtle changes in audio waveforms.

[0021] The sound wave feature extraction module is used to perform feature analysis on audio waveforms to obtain the sound wave features of thermal paper; The sound wave feature extraction module is configured with a sound wave feature extraction strategy, which includes: Please see Figure 3 As shown, the audio waveform is stored in the audio graph. The X-axis of the audio graph represents time, and the Y-axis represents amplitude. The coordinates of the peak of the audio waveform are obtained and named as peak points, and the coordinates of the trough of the audio waveform are obtained and named as trough points. The peaks and troughs are numbered in chronological order, using the symbol F. n and G n This represents the expression, where n is a non-zero natural number and n is the index of F and G. Typically, the number of peaks and valleys is the same, and G... n For located at F n The first valley after that; Please see Figure 4 As shown, F is connected by a straight line. n With G n Connect the lines and mark the resulting auxiliary lines as L. n , obtain L n The slope, denoted as K n ; Please see Figure 5 As shown, through a straight line, two adjacent F n Connecting the segments yields a single line segment, which is named the peak analysis line. In practical applications, because the audio waveforms in the original audio graph are too dense to be observed visually, peaks and troughs are only shown as examples in this embodiment to illustrate the subsequent analysis process; the audio waveforms shown in this embodiment are as follows: Figure 3 As shown, both peaks and valleys are marked, and F is obtained by numbering them. n and G n Where 1≤n≤4, the connection yields L. n like Figure 4 As shown, Figure 4 The dashed line in the middle is L. n Get K n Then connect them to obtain the peak analysis polyline, as shown below. Figure 5 As shown.

[0022] Please see Figure 6 As shown, F n The corresponding time stamp is T n , will K n With T n Related, with T n K is the horizontal axis. n Establish a Cartesian coordinate system for the vertical axis, named the Peak-Valley Analysis Chart, and set K... n According to T n Enter the peak-valley analysis chart and name the coordinate points in the chart as peak-valley analysis points; K n With T n The coordinate point formed is labeled P. n Through a straight line to two adjacent P n Connecting the segments yields a broken line segment, which is named the Peak-Valley Analysis Broken Line. The peak analysis line and the trough analysis line are the characteristics of sound waves; In practical applications, the peak-valley analysis diagram is constructed as follows: Figure 6 As shown, Figure 6 The diagram also provides a broken line for peak and valley analysis.

[0023] The feature analysis module is used to build an online detection model for the tear resistance of thermal paper, collect the surface features and acoustic features of thermal paper of known quality and perform deep learning; the feature analysis module includes a surface analysis unit and an audio analysis unit; The surface analysis unit is used to build an online detection model for the tear resistance of thermal paper, collect the surface features of thermal paper of known quality and perform deep learning to obtain surface reference features; The surface analysis unit is configured with surface analysis strategies, which include: An online detection model for the tear resistance of thermal paper was constructed. Thermal paper of known quality was used as sample paper. The quality of the sample paper was named as sample quality. The surface features of the sample paper were extracted and named as surface sample features. The sample quality corresponds to the surface sample features. The contrast and correlation in the surface sample features are named sample contrast and sample correlation, respectively. The range of contrast of samples with acceptable sample quality is used to obtain the first reference range, and the range of correlation of samples with acceptable sample quality is used to obtain the second reference range. The first reference range and the second reference range together constitute the surface reference feature. In practical applications, contrast reflects the clarity of an image and the depth of its texture, while correlation reflects the similarity of gray levels in the row or column direction. Both directly reflect the characteristics of the thermal paper surface. When the tear resistance is unqualified, it indicates a defect on the thermal paper surface. This defect will cause significant changes in correlation and contrast. Therefore, their ranges can be directly calculated to obtain the first and second reference ranges. If the contrast is within the first reference range and the correlation is within the second reference range, it means that the thermal paper surface meets the characteristics of thermal paper with normal tear resistance. However, in special cases, machine vision cannot detect changes in tear resistance. Therefore, machine vision solutions have extremely high requirements for light sources and imaging equipment. Even slight deviations can lead to errors in the detection results. Therefore, this embodiment adds audio waveform recognition to verify whether the tear resistance of the thermal paper is normal.

[0024] The audio analysis unit is used to collect the acoustic wave characteristics of thermal paper of known quality and perform deep learning to obtain acoustic wave reference characteristics; The audio analysis unit is configured with audio analysis strategies, which include: Acoustic characteristics of sample paper with qualified sample quality were collected and named as acoustic sample characteristics. The peak analysis line and the peak-valley analysis line in the acoustic sample characteristics were named as peak sample line and peak-valley sample line, respectively. Name the leftmost endpoint of the peak sample line as the peak start point, name the leftmost endpoint of the valley sample line as the valley start point, and name the turning points and endpoints in the peak and valley sample lines as line vertices. Please see Figure 7 As shown, the peak starting points of all peak sample lines are overlapped, and the overlapping image is named the peak learning image; the peak and valley starting points of all valley sample lines are overlapped, and the overlapping image is named the valley learning image. When performing deep learning on peak learning graphs or valley learning graphs, the peak learning graph or valley learning graph being analyzed is named the target analysis graph, and the X-axis and Y-axis corresponding to the target analysis graph are named the first axis and the second axis, respectively. In practical applications, since the analysis process for both peak and trough sample line graphs is identical, this embodiment only uses the analysis process of the peak sample line graph as an example to illustrate the subsequent analysis process. Due to the large amount of data, it is inconvenient to observe in this embodiment; therefore, this embodiment only uses a small amount of data as an example to construct the peak learning graph as shown below. Figure 7 As shown, Figure 7 That is, the target analysis diagram. Figure 7 The X-axis is the first axis, and the Y-axis is the second axis.

[0025] Please see Figure 8 As shown, for each value of the first axis, the topmost and bottommost polyline vertices are obtained and named the upper limit point and lower limit point, respectively. The upper and lower limits are numbered from left to right, and are respectively represented by the symbol U. i and D j This indicates that i and j are both non-zero natural numbers, i is the index of U, and j is the index of D; Set an auxiliary number h, initially 1, starting with i=1, and pass through U via a straight line. i with U i+h Connect the points to obtain the upper limit analysis line. Determine whether the upper limit analysis line intersects with any part of the target analysis diagram except for the upper limit point. If yes, output an upper limit error signal; otherwise, output an upper limit correct signal. If an upper limit error signal is output, the upper limit analysis line will be deleted, and U will be... i+h Delete, increment h and reanalyze the upper limit analysis line, repeat the judgment until the upper limit is correctly output; if the upper limit is correctly output, retain the upper limit analysis line, and increment i by 1, then judge U. i Does it exist, if Ui If it does not exist, increment i again; if U i If it exists, reset h to 1 and analyze another upper limit analysis line again until the maximum value of i is reached, to obtain different upper limit analysis lines. The upper limit analysis lines form a new polyline segment, which is named the upper limit reference line. In practical applications, the upper and lower limits are obtained as follows: Figure 8 As shown, U is obtained through numbering. i and D j In this context, i is actually equal to j, and 1 ≤ i = j ≤ 4. An auxiliary number h is set, initially set to 1. When i = 1, U1 and U2 are connected to obtain the upper limit analysis line. Within the range of the first axis where the upper limit analysis line is located, if all values ​​in the peak learning graph are less than or equal to the upper and lower analysis lines, then the upper limit correct signal is output. In fact, it means that all broken lines within the range of the first axis where the upper limit analysis line is located are not above the upper limit analysis line. Under normal circumstances, the upper limit error signal will not be output. Setting the upper limit error signal is only to prevent errors from occurring during the analysis process, so that errors can be detected in time and the analysis can be re-performed.

[0026] Reset h to 1, starting with j=1, and move D through a straight line. j With D j+h Connect the points to obtain the lower limit analysis line. Determine whether the lower limit analysis line intersects with any part of the target analysis diagram except for the lower limit point. If yes, output the lower limit error signal; otherwise, output the lower limit correct signal. If a lower limit error signal is output, the lower limit analysis line will be deleted, and D will be... j+h Delete, increment h and re-analyze the lower limit analysis line, repeat the judgment until the lower limit correct signal is output; if the lower limit correct signal is output, retain the lower limit analysis line, and increment j by 1, then judge D. j Does it exist if D j If it does not exist, increment j again. If D j If it exists, reset h to 1 and analyze another lower limit analysis line again until the maximum value of j is reached, to obtain different lower limit analysis lines. The lower limit analysis lines form a new line segment, which is named the lower limit reference line. Please see Figure 9 As shown, the leftmost endpoints of the upper limit reference line and the lower limit reference line are named the upper limit left endpoint and the lower limit left endpoint, respectively. The rightmost endpoints of the upper limit reference line and the lower limit reference line are named the upper limit right endpoint and the lower limit right endpoint, respectively. The upper limit left endpoint and the lower limit left endpoint are connected, and the upper limit right endpoint and the lower limit right endpoint are connected to obtain a closed region, which is named the target reference feature. The target reference features obtained from the analysis of the peak learning map and the peak-valley learning map are named peak reference features and peak-valley reference features, respectively. The peak reference features and peak-valley reference features together constitute the sound wave reference features. In practical applications, the process of analyzing the lower limit reference line is exactly the same as that of analyzing the upper limit reference line. This embodiment will not provide further details. The final analysis yields the peak reference characteristics as follows: Figure 9 As shown, Figure 9 The gray area in the image represents the peak reference feature.

[0027] The tear resistance strength assessment module is used to analyze surface features and acoustic features through an online tear resistance strength detection model for thermal paper to determine whether the tear resistance strength of the thermal paper is qualified. The tear strength assessment module is configured with a tear strength assessment strategy, which includes: The surface features and acoustic features to be analyzed are named Real-time Surface Features and Real-time Acoustic Features, respectively. The contrast and correlation in the Real-time Surface Features are named Real-time Contrast and Real-time Correlation, respectively. The peak analysis line and peak-valley analysis line in the Real-time Acoustic Features are named Real-time Peak Line and Real-time Peak-valley Line, respectively. If the real-time contrast is within the first reference range, the real-time correlation is within the second reference range, and all real-time peak lines are within the peak reference features and all real-time trough lines are within the trough reference features, then the tear resistance strength is qualified; otherwise, the tear resistance strength is unqualified. In practical applications, the process for judging tear strength has been clearly explained in the tear strength judgment strategy. Only when all conditions are met can the tear strength be judged to be qualified. This embodiment will not be described in detail.

[0028] Example 2, please refer to Figure 10 As shown, this application provides an online detection method for the tear resistance of thermal paper based on machine vision, including the following steps: Step S1 involves installing a light source and a camera on the thermal paper production line to capture images of the paper. Step S1 includes the following sub-steps: Step S101: The light source is a flicker-free LED linear light source, and the camera is a line scan camera; In step S102, the LED linear light source illuminates the thermal paper at a first angle, and the line scan camera is positioned directly above the thermal paper to take a picture of the thermal paper, thereby obtaining an image of the paper. Step S2 involves installing a paper tapping device on the thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped, naming it an audio waveform. Step S2 includes the following sub-steps: Step S201, the paper tapping device includes a front conveyor belt, a rear conveyor belt, a front fixing device, a rear fixing device, and a tapping device; Step S202: The front fixing device is installed above the front conveyor belt, and the rear fixing device is installed above the rear conveyor belt. In step S203, before the thermal paper is tapped, the front conveyor belt stops running, the front fixing device presses down to fix the front end of the thermal paper, the rear conveyor belt stops running, the rear fixing device presses down to fix the rear end of the thermal paper, and then the rear conveyor belt is started. The rear fixing device can move synchronously with the rear conveyor belt. At this time, a protrusion will be generated on the thermal paper between the front fixing device and the rear fixing device. The tapping device taps the protrusion and collects the waveform of the sound generated by the thermal paper during the tapping to obtain the audio waveform. Step S3 involves performing feature analysis on the paper image to extract the surface features of the thermal paper. Step S3 includes the following sub-steps: Step S301: Perform grayscale processing on the paper image to obtain a grayscale image; Step S302: Obtain the gray-level co-occurrence matrix of the gray-level image, and extract the contrast and correlation in the gray-level co-occurrence matrix as the surface features of the thermal paper. Step S4 involves performing feature analysis on the audio waveform to obtain the acoustic characteristics of the thermal paper. Step S4 includes the following sub-steps: Step S401: The audio waveform is stored in the audio graph. The X-axis of the audio graph represents time, and the Y-axis represents amplitude. The coordinates of the peak of the audio waveform are obtained and named as peak points. The coordinates of the trough of the audio waveform are obtained and named as trough points. Step S402: Number the peaks and valleys according to the chronological order, and use the symbol F respectively. n and G n This represents the expression, where n is a non-zero natural number and n is the index of F and G. Typically, the number of peaks and valleys is the same, and G... n For located at F n The first valley after that; Step S403, pass F through a straight line n With G n Connect the lines and mark the resulting auxiliary lines as L. n , obtain L n The slope, denoted as K n ; Step S404, pass through two adjacent Fs by a straight line n Connecting the segments yields a single line segment, which is named the peak analysis line. Step S405, F n The corresponding time stamp is T n , will K nWith T n Related, with T n K is the horizontal axis. n Establish a Cartesian coordinate system for the vertical axis, named the Peak-Valley Analysis Chart, and set K... n According to T n Enter the peak-valley analysis chart and name the coordinate points in the chart as peak-valley analysis points; Step S406, K n With T n The coordinate point formed is labeled P. n Through a straight line to two adjacent P n Connecting the segments yields a broken line segment, which is named the Peak-Valley Analysis Broken Line. Step S407, the peak analysis line and the peak-valley analysis line are the sound wave characteristics; Step S5: Construct an online detection model for the tear resistance of thermal paper, collect surface features and acoustic features of thermal paper of known quality, and perform deep learning; Step S5 includes the following sub-steps: Step S501: Construct an online detection model for the tear resistance of thermal paper, collect surface features of thermal paper of known quality and perform deep learning to obtain surface reference features; Step S501 includes the following sub-steps: Step S501.01: Construct an online detection model for the tear resistance of thermal paper, obtain thermal paper of known quality as sample paper, name the quality of the sample paper as sample quality, extract the surface features of the sample paper and name them as surface sample features, and the sample quality corresponds to the surface sample features. Step S501.02: Name the contrast and correlation in the surface sample features as sample contrast and sample correlation, respectively. Step S501.03: Calculate the range of contrast of samples with acceptable sample quality to obtain the first reference range; calculate the range of correlation of samples with acceptable sample quality to obtain the second reference range; the first reference range and the second reference range together constitute the surface reference feature. Step S502: Collect the acoustic wave characteristics of thermal paper of known quality and perform deep learning to obtain acoustic wave reference characteristics; Step S502 includes the following sub-steps: Step S502.01: Collect the acoustic characteristics of the sample paper with qualified sample quality, name it acoustic sample characteristics, and name the peak analysis line and the valley analysis line in the acoustic sample characteristics as peak sample line and valley sample line, respectively. Step S502.02: Name the leftmost endpoint of the peak sample polyline as the peak start point, name the leftmost endpoint of the valley sample polyline as the valley start point, and name the turning points and endpoints in the peak sample polyline and valley sample polyline as the polyline vertices. Step S502.03: Overlap the peak start points of all peak sample polylines and name the overlapping image as a peak learning graph; overlap the peak and valley start points of all valley sample polylines and name the overlapping image as a valley learning graph. Step S502.04: When performing deep learning on the peak learning map or the valley learning map, name the currently analyzed peak learning map or valley learning map the target analysis map, and name the X-axis and Y-axis corresponding to the target analysis map the first axis and the second axis, respectively. Step S502.05: For each value of the first axis, obtain the topmost and bottommost polyline vertices and name them the upper limit point and lower limit point, respectively. Step S502.06: Number the upper limit point and lower limit point in order from left to right, and use the symbol U respectively. i and D j This indicates that i and j are both non-zero natural numbers, i is the index of U, and j is the index of D; Step S502.07: Set the auxiliary number h, initially 1, starting with i=1, and use a straight line to connect U... i with U i+h Connect the points to obtain the upper limit analysis line. Determine whether the upper limit analysis line intersects with any part of the target analysis diagram except for the upper limit point. If yes, output an upper limit error signal; otherwise, output an upper limit correct signal. Step S502.08: If an upper limit error signal is output, delete the upper limit analysis line and simultaneously set U... i+h Delete, increment h and reanalyze the upper limit analysis line, repeat the judgment until the upper limit is correctly output; if the upper limit is correctly output, retain the upper limit analysis line, and increment i by 1, then judge U. i Does it exist, if U i If it does not exist, increment i again; if U i If it exists, reset h to 1 and analyze another upper limit analysis line again until the maximum value of i is reached, to obtain different upper limit analysis lines. The upper limit analysis lines form a new polyline segment, which is named the upper limit reference line. Step S502.09: Reset h to 1, starting with j=1, and move D through a straight line. j With D j+h Connect the points to obtain the lower limit analysis line. Determine whether the lower limit analysis line intersects with any part of the target analysis diagram except for the lower limit point. If yes, output the lower limit error signal; otherwise, output the lower limit correct signal. Step S502.10: If a lower limit error signal is output, delete the lower limit analysis line and simultaneously set D... j+h Delete, increment h and re-analyze the lower limit analysis line, repeat the judgment until the lower limit correct signal is output; if the lower limit correct signal is output, retain the lower limit analysis line, and increment j by 1, then judge D. j Does it exist if D j If it does not exist, increment j again. If D j If it exists, reset h to 1 and analyze another lower limit analysis line again until the maximum value of j is reached, to obtain different lower limit analysis lines. The lower limit analysis lines form a new line segment, which is named the lower limit reference line. Step S502.11: Name the leftmost endpoints of the upper limit reference line and the lower limit reference line as the upper limit left endpoint and the lower limit left endpoint, respectively. Name the rightmost endpoints of the upper limit reference line and the lower limit reference line as the upper limit right endpoint and the lower limit right endpoint, respectively. Connect the upper limit left endpoint and the lower limit left endpoint, and connect the upper limit right endpoint and the lower limit right endpoint to obtain a closed region, which is named the target reference feature. Step S502.12: Name the target reference features obtained from the analysis of the peak learning map and the valley learning map as peak reference features and valley reference features, respectively. The peak reference features and valley reference features together constitute the sound wave reference features. Step S6 involves analyzing surface and acoustic characteristics using an online tear resistance testing model for thermal paper to determine whether the tear resistance of the thermal paper is up to standard. Step S6 includes the following sub-steps: Step S601: Name the surface features and acoustic features to be analyzed as real-time surface features and real-time acoustic features, respectively. Name the contrast and correlation in the real-time surface features as real-time contrast and real-time correlation, respectively. Name the peak analysis line and peak-valley analysis line in the real-time acoustic features as real-time peak line and real-time peak-valley line, respectively. Step S602: If the real-time contrast is within the first reference range, the real-time correlation is within the second reference range, and all real-time peak lines are within the peak reference features, and all real-time peak and valley lines are within the peak and valley reference features, then output a tear resistance strength qualified signal; otherwise, output a tear resistance strength unqualified signal.

[0029] Example 3: This application provides an electronic device that may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps such as those in the online detection method for the tear resistance of thermal paper based on machine vision, to achieve the following functions: acquiring a paper image; recording the audio waveform generated when the thermal paper is struck; extracting the surface features of the thermal paper from the paper image; obtaining the acoustic wave features of the thermal paper; constructing an online detection model for the tear resistance of thermal paper; acquiring the surface features and acoustic wave features of thermal paper of known quality and performing deep learning; and analyzing the surface features and acoustic wave features using the online detection model for the tear resistance of thermal paper to determine whether the tear resistance of the thermal paper is qualified.

[0030] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0031] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-described online detection method for the tear resistance of thermal paper based on machine vision, to achieve the following functions: acquiring a paper image; recording the audio waveform generated when the thermal paper is tapped; extracting the surface features of the thermal paper from the paper image; obtaining the acoustic features of the thermal paper; constructing an online detection model for the tear resistance of thermal paper; acquiring the surface features and acoustic features of thermal paper of known quality and performing deep learning; and analyzing the surface features and acoustic features through the online detection model for the tear resistance of thermal paper to determine whether the tear resistance of the thermal paper is qualified.

[0032] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0033] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for online detection of the tear resistance of thermal paper based on machine vision, characterized in that, Includes the following steps: Light sources and cameras are installed on the thermal paper production line to capture images of the paper. A paper tapping device is installed on the thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped, which is named audio waveform. Feature analysis is performed on paper images to extract the surface features of thermal paper from the paper images; Feature analysis of audio waveforms is performed to obtain the sound wave characteristics of thermal paper; An online detection model for the tear resistance of thermal paper was constructed, and the surface and acoustic features of thermal paper of known quality were collected and deep learning was performed. The surface and acoustic characteristics of thermal paper are analyzed using an online tear resistance testing model to determine whether the tear resistance of the thermal paper is up to standard.

2. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 1, characterized in that, Installing a light source and camera on a thermal paper production line to capture images of the paper includes the following sub-steps: The light source is a flicker-free LED linear light source, and the camera is a line scan camera; The LED linear light source illuminates the thermal paper at a first angle, and the linear array camera is positioned directly above the thermal paper to capture an image of the paper.

3. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 2, characterized in that, Installing a paper tapping device on a thermal paper production line to tap the thermal paper and record the waveform of the sound generated when the thermal paper is tapped includes the following sub-steps: The paper tapping device includes a front conveyor belt, a rear conveyor belt, a front fixing device, a rear fixing device, and a tapping device. The front fixing device is installed above the front conveyor belt, and the rear fixing device is installed above the rear conveyor belt. Before the thermal paper is tapped, the front conveyor belt stops running, the front fixing device presses down to fix the front end of the thermal paper, the rear conveyor belt stops running, the rear fixing device presses down to fix the rear end of the thermal paper, and then the rear conveyor belt is started. The rear fixing device can move synchronously with the rear conveyor belt. At this time, a protrusion will be generated on the thermal paper between the front fixing device and the rear fixing device. The tapping device taps the protrusion and collects the waveform of the sound generated by the thermal paper during the tapping to obtain the audio waveform.

4. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 3, characterized in that, Feature analysis of paper images to extract surface features of thermal paper includes the following sub-steps: The paper image is converted to grayscale to obtain a grayscale image; Obtain the gray-level co-occurrence matrix of the gray-level image, and extract the contrast and correlation from the gray-level co-occurrence matrix as surface features of the thermal paper.

5. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 4, characterized in that, The process of performing feature analysis on audio waveforms to obtain the acoustic characteristics of thermal paper includes the following sub-steps: The audio waveform is stored in an audio graph, where the X-axis represents time and the Y-axis represents amplitude. The coordinates of the peaks of the audio waveform are obtained and named peak points, and the coordinates of the troughs of the audio waveform are obtained and named trough points. The peaks and troughs are numbered in chronological order, using the symbol F. n and G n This represents the expression, where n is a non-zero natural number and n is the index of F and G. Typically, the number of peaks and valleys is the same, and G... n For located at F n The first valley after that; F through the straight line n With G n Connect the lines and mark the resulting auxiliary lines as L. n , obtain L n The slope, denoted as K n ; Through a straight line to two adjacent F n Connecting the segments yields a single line segment, which is named the peak analysis line. F n The corresponding time stamp is T n , will K n With T n Related, with T n K is the horizontal axis. n Establish a Cartesian coordinate system for the vertical axis, named the Peak-Valley Analysis Chart, and set K... n According to T n Enter the peak-valley analysis chart and name the coordinate points in the chart as peak-valley analysis points; K n With T n The coordinate point formed is labeled P. n Through a straight line to two adjacent P n Connecting the segments yields a broken line segment, which is named the Peak-Valley Analysis Broken Line. The peak analysis line and the peak-valley analysis line are the characteristics of the sound wave.

6. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 5, characterized in that, Constructing an online detection model for the tear resistance of thermal paper involves collecting surface and acoustic features of thermal paper of known quality and performing deep learning, including the following sub-steps: An online detection model for the tear resistance of thermal paper was constructed. The surface features of thermal paper of known quality were collected and deep learning was performed to obtain surface reference features. Acoustic wave characteristics of thermal paper of known quality are collected and deep learning is performed to obtain acoustic wave reference characteristics.

7. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 6, characterized in that, To construct an online detection model for the tear resistance of thermal paper, surface features of thermal paper of known quality are collected and deep learning is performed to obtain surface reference features, including the following sub-steps: An online detection model for the tear resistance of thermal paper is constructed. Thermal paper of known quality is used as sample paper. The quality of the sample paper is named as sample quality. The surface features of the sample paper are extracted and named as surface sample features. The sample quality corresponds to the surface sample features. The contrast and correlation in the surface sample features are named sample contrast and sample correlation, respectively. The range of sample contrast that represents acceptable sample quality is used to obtain a first reference range, and the range of sample correlation that represents acceptable sample quality is used to obtain a second reference range. The first reference range and the second reference range together constitute the surface reference feature.

8. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 7, characterized in that, The process of collecting acoustic wave characteristics of thermal paper of known quality and performing deep learning to obtain acoustic wave reference characteristics includes the following sub-steps: Acoustic characteristics of sample paper with qualified sample quality were collected and named as acoustic sample characteristics. The peak analysis line and the peak-valley analysis line in the acoustic sample characteristics were named as peak sample line and peak-valley sample line, respectively. Name the leftmost endpoint of the peak sample line as the peak start point, name the leftmost endpoint of the valley sample line as the valley start point, and name the turning points and endpoints in the peak and valley sample lines as line vertices. Overlay the peak starting points of all peak sample polylines and name the overlaid image the peak learning graph. Overlay the peak and valley starting points of all peak and valley sample polylines, and name the overlaid image a peak and valley learning graph. When performing deep learning on peak learning graphs or valley learning graphs, the peak learning graph or valley learning graph being analyzed is named the target analysis graph, and the X-axis and Y-axis corresponding to the target analysis graph are named the first axis and the second axis, respectively. For each value of the first axis, obtain the topmost and bottommost polyline vertices and name them the upper limit point and lower limit point, respectively. The upper and lower limits are numbered from left to right, and are respectively represented by the symbol U. i and D j This indicates that i and j are both non-zero natural numbers, i is the index of U, and j is the index of D; Set an auxiliary number h, which is initially 1. Starting with i=1, pass through the straight line U i with U i+h Connect the points to obtain the upper limit analysis line. Determine whether the upper limit analysis line intersects with any part of the target analysis diagram except for the upper limit point. If yes, output an upper limit error signal; otherwise, output an upper limit correct signal. If an upper limit error signal is output, the upper limit analysis line will be deleted, and U will be... i+h Delete, increment h and reanalyze the upper limit analysis line, repeat the judgment until the upper limit is correctly output; if the upper limit is correctly output, retain the upper limit analysis line, and increment i by 1, then judge U. i Does it exist, if U i If it does not exist, increment i again; if U i If it exists, reset h to 1 and analyze another upper limit analysis line again until the maximum value of i is reached, to obtain different upper limit analysis lines. The upper limit analysis lines form a new polyline segment, which is named the upper limit reference line. Reset h to 1, starting with j=1, and move D through a straight line. j With D j+h Connect the points to obtain the lower limit analysis line. Determine whether the lower limit analysis line intersects with any part of the target analysis diagram except for the lower limit point. If yes, output the lower limit error signal; otherwise, output the lower limit correct signal. If a lower limit error signal is output, the lower limit analysis line will be deleted, and D will be... j+h Delete, increment h and re-analyze the lower limit analysis line, repeat the judgment until the lower limit correct signal is output; if the lower limit correct signal is output, retain the lower limit analysis line, and increment j by 1, then judge D. j Does it exist if D j If it does not exist, increment j again. If D j If it exists, reset h to 1 and analyze another lower limit analysis line again until the maximum value of j is reached, to obtain different lower limit analysis lines. The lower limit analysis lines form a new line segment, which is named the lower limit reference line. Name the leftmost endpoints of the upper and lower reference lines as the upper left endpoint and the lower left endpoint, respectively. Name the rightmost endpoints of the upper and lower reference lines as the upper right endpoint and the lower right endpoint, respectively. Connect the upper left endpoint and the lower left endpoint, and connect the upper right endpoint and the lower right endpoint to obtain a closed region, which is named the target reference feature. The target reference features obtained from the analysis of the peak learning map and the peak-valley learning map are named peak reference features and peak-valley reference features, respectively. The peak reference features and peak-valley reference features together constitute the sound wave reference features.

9. The online detection method for the tear resistance of thermal paper based on machine vision according to claim 8, characterized in that, The online tear resistance testing model for thermal paper analyzes surface and acoustic characteristics to determine whether the tear resistance of the thermal paper is qualified. The process includes the following sub-steps: The surface features and acoustic features to be analyzed are named Real-time Surface Features and Real-time Acoustic Features, respectively. The contrast and correlation in the Real-time Surface Features are named Real-time Contrast and Real-time Correlation, respectively. The peak analysis line and peak-valley analysis line in the Real-time Acoustic Features are named Real-time Peak Line and Real-time Peak-valley Line, respectively. If the real-time contrast is within the first reference range, the real-time correlation is within the second reference range, and all real-time peak lines are within the peak reference features and all real-time trough lines are within the trough reference features, then the tear resistance strength is qualified; otherwise, the tear resistance strength is unqualified.

10. A machine vision-based online tear resistance testing system for thermal paper, used to implement the machine vision-based online tear resistance testing method for thermal paper as described in any one of claims 1-9, characterized in that, It includes an image acquisition module, an audio acquisition module, a surface feature extraction module, a sound wave feature extraction module, a feature analysis module, and a tear resistance strength determination module; the image acquisition module, audio acquisition module, surface feature extraction module, sound wave feature extraction module, and tear resistance strength determination module are all data connected to the feature analysis module; The image acquisition module is used to install a light source and a camera on the thermal paper production line to acquire images of the paper. The audio acquisition module is used to install a paper tapping device on the thermal paper production line, which taps the thermal paper and records the waveform of the sound generated when the thermal paper is tapped, named audio waveform. The surface feature extraction module is used to perform feature analysis on the paper image and extract the surface features of the thermal paper in the paper image; The sound wave feature extraction module is used to perform feature analysis on the audio waveform to obtain the sound wave features of the thermal paper; The feature analysis module is used to construct an online detection model for the tear resistance of thermal paper, collect the surface features and acoustic features of thermal paper of known quality and perform deep learning; The tear resistance strength judgment module is used to analyze surface features and acoustic features through an online detection model for the tear resistance strength of thermal paper to determine whether the tear resistance strength of the thermal paper is qualified.

Images