Mica pulp quality control method, medium and system based on image recognition

Abstract

This application discloses a mica pulp quality control method, medium, and system based on image recognition, relating to the field of image recognition. The method includes: obtaining an image of a target area of ​​mica pulp by operating a vibrator at the bottom of a mica pulp pool; performing image recognition on the target area image to obtain vibration ripple characteristics of the target area image; determining a first concentration range of the mica pulp based on the vibration ripple characteristics of the target area image; emitting incident light to the target area; obtaining a current image of the target area; determining the number of scales on the surface of the mica pulp based on the current image; determining a second concentration range of the mica pulp based on the number of scales on the mica pulp surface; and determining a final concentration range of the mica pulp based on the first and second concentration ranges of the mica pulp, thereby generating a quality control strategy based on the final concentration range. This application can effectively improve the accuracy of mica pulp quality control.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image recognition, and in particular to a mica pulp quality control method, medium, and system based on image recognition. Background Art

[0002] Mica paper has the characteristics of electrical insulation and thermal insulation, and is widely used in electrical, construction, aerospace, chemical, energy and other fields. The quality control of mica pulp is a crucial step in the mica paper production process.

[0003] Existing technology primarily controls mica pulp quality through manual inspection. During the mica pulp production process, mica pulp concentration and impurities are manually observed to ensure quality. However, manual inspection using the naked eye can lead to visual fatigue and subjective analysis of mica pulp quality. These factors can affect the effectiveness of mica pulp quality inspections and result in poor quality of the finished mica paper.

[0004] Therefore, there is an urgent need for a mica pulp quality control method, medium and system based on image recognition. Summary of the Invention

[0005] The embodiments of the present application provide a mica pulp quality control method, medium, and system based on image recognition, which are used to solve the problem of inaccurate mica pulp quality detection.

[0006] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0007] In a first aspect, a mica pulp quality control method based on image recognition is provided, the method comprising:

[0008] issuing a first control instruction to control the operation of a vibrator located at the bottom of the mica pulp tank;

[0009] Acquiring a target area image of mica pulp, performing image recognition on the target area image, and obtaining vibration ripple features of the target area image, wherein the vibration ripple features include the number of vibration ripples and the amplitude of the vibration ripples;

[0010] determining a first concentration range of mica pulp according to the vibration ripple characteristics of the target area image, and judging whether the first concentration range is within a preset concentration range;

[0011] If the first concentration range is not within the preset concentration range, issuing an early warning instruction;

[0012] If the first concentration range is within a preset concentration range, after the mica pulp is allowed to stand for a preset period of time, a second control instruction is issued to control the light source to emit incident light toward a target area;

[0013] Acquire a current image of the target area;

[0014] determining the number of scales on the surface of the mica pulp using the current image;

[0015] determining a second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp;

[0016] The first concentration range and the second concentration range of the mica pulp are combined to determine a final concentration range of the mica pulp, so as to generate a quality control strategy according to the final concentration range.

[0017] In a possible implementation of the first aspect, performing image recognition on the target area image to obtain the vibration ripple characteristics of the target area includes:

[0018] Dividing the target area image into a plurality of unit grid images, and preprocessing each of the unit grid images to obtain a corresponding grayscale unit grid image;

[0019] Image recognition is performed on each of the grayscale unit grid images to obtain the number of vibration ripples and the amplitude of the vibration ripples.

[0020] In another possible implementation manner of the first aspect, after preprocessing each of the unit grid images to obtain a corresponding grayscale unit grid image, the method further includes:

[0021] For any one of the grayscale unit grid images, performing denoising processing on the grayscale unit grid image;

[0022] Performing feature extraction on the grayscale unit grid image after denoising to obtain target impurity edge features;

[0023] Acquire a first pixel point of the target impurity edge feature and a second pixel point of the mica flake edge feature;

[0024] Calculating the Euclidean distance and color contrast between the first pixel and the second pixel, and determining whether the target area is a contaminated area based on the Euclidean distance and color contrast between the first pixel and the second pixel;

[0025] If the Euclidean distance and the color contrast are both greater than a preset threshold, the target area is determined to be a polluted area and an alarm message is issued;

[0026] If the Euclidean distance and the color contrast are both less than or equal to a preset threshold, it is determined that the target area is not a contaminated area, and the step of performing image recognition on each grayscale unit grid image to obtain the number of vibration ripples and the amplitude of the vibration ripples is performed.

[0027] In another possible implementation of the first aspect, determining the first concentration range of the mica pulp according to the vibration ripple characteristics of the target area includes:

[0028] Obtaining the number of vibration ripples and the amplitude of vibration ripples according to the vibration ripple characteristics of the target area;

[0029] Comparing the number of vibration ripples with the number of vibration ripples of a preset concentration to obtain a first concentration of mica pulp;

[0030] Comparing the vibration ripple amplitude with the vibration ripple amplitude of a preset concentration to obtain a second concentration of the mica pulp;

[0031] The first concentration range of the mica pulp is determined by combining the first concentration of the mica pulp and the second concentration of the mica pulp.

[0032] In another possible implementation of the first aspect, if the first concentration range is not within a preset concentration range, issuing a warning instruction for mica pulp concentration includes:

[0033] comparing the first concentration range with a preset concentration range to determine a warning level for the first concentration range;

[0034] Based on the warning level, a warning instruction for mica pulp concentration is issued.

[0035] In another possible implementation of the first aspect, determining the number of scales on the surface of the mica pulp using the current image includes:

[0036] Performing grayscale processing on the current image to obtain a grayscale image, wherein the current image includes a plurality of reflected light points obtained based on the incident light;

[0037] Acquire a third pixel point of the grayscale image and a fourth pixel point of the background image of the mica pulp surface; acquire pixel points of the current image, and identify color features of all the pixel points of the current image;

[0038] Determine the pixel point of the reflected light point according to the color feature;

[0039] According to the pixel points of the reflected light spots, a preset edge detection algorithm is used to determine the edges of the reflected light spots, and based on the edges of the reflected light spots, the number of the reflected light spots is obtained, wherein the number of the reflected light points is used to characterize the number of scales on the surface of the mica pulp.

[0040] In another possible implementation of the first aspect, determining the second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp includes:

[0041] The number of scales on the surface of the mica pulp is retrieved in a preset database to obtain a second concentration range of the mica pulp corresponding to the number of scales on the surface of the mica pulp.

[0042] In another possible implementation of the first aspect, determining a final concentration range of the mica pulp by combining the first concentration range and the second concentration range of the mica pulp includes:

[0043] When the second concentration range of the mica pulp is included in the first concentration range, the second concentration range is set as the final concentration range of the mica pulp;

[0044] When the first concentration range and the second concentration range of the mica pulp overlap, the overlapping range of the first concentration range and the second concentration range is used as the final concentration range of the mica pulp;

[0045] When the first concentration range and the second concentration range of the mica pulp are adjacent to each other, the range obtained by adding the first concentration range and the second concentration range is used as the final concentration range of the mica pulp.

[0046] In a second aspect, the present application provides a machine-readable storage medium having stored thereon instructions for causing a machine to execute the above-mentioned mica pulp quality control method based on image recognition.

[0047] In a third aspect, the present application provides a mica pulp quality control system based on image recognition, comprising:

[0048] a memory configured to store instructions; and

[0049] The processor is configured to call the instructions from the memory and implement the above-mentioned mica pulp quality control method based on image recognition when executing the instructions.

[0050] The above technical solution uses a vibrator at the bottom of the mica pulp tank to vibrate the mica pulp, capturing a target area on the mica pulp surface. The vibration ripple characteristics of the target area image are then determined using image recognition. This allows for a more precise quality control of the mica pulp concentration. Based on the vibration ripple characteristics, a first concentration range of the mica pulp is determined. Obtaining the first concentration range facilitates more accurate concentration analysis of the mica pulp, enhancing mica pulp quality control. If the first concentration range falls outside the preset concentration range, an early warning is issued, alerting personnel and helping to promptly identify potential risks associated with the mica pulp. By determining the number of scales on the mica pulp surface, a second concentration range of the mica pulp is determined, and a final concentration range is determined. A quality control strategy is generated based on the final concentration range, enabling precise control of the mica pulp concentration, effectively ensuring accurate concentration during production.

[0051] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic flow chart of a mica pulp quality control method based on image recognition provided in an embodiment of the present application;

[0053] Figure 2 A schematic diagram of a process for detecting contaminated areas in a target area provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0055] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), such directional indications are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0056] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0057] Figure 1 The following schematically shows a flow chart of a mica pulp quality control method based on image recognition according to an embodiment of the present application. Figure 1 As shown, an embodiment of the present application provides a mica pulp quality control method based on image recognition, which may include the following steps.

[0058] S110, issuing a first control instruction to control the operation of a vibrator located at the bottom of the mica pulp pool;

[0059] S120, acquiring a target area image of mica pulp, performing image recognition on the target area image, and obtaining vibration ripple features of the target area image, wherein the vibration ripple features include the number of vibration ripples and the amplitude of the vibration ripples;

[0060] S130, determining a first concentration range of the mica pulp based on the vibration ripple characteristics of the target area image, and determining whether the first concentration range is within a preset concentration range;

[0061] S140: If the first concentration range is not within the preset concentration range, issue a warning instruction;

[0062] S150: If the first concentration range is within the preset concentration range, after the mica pulp is allowed to stand for a preset period of time, issuing a second control instruction to control the light source to emit incident light toward the target area;

[0063] S160, obtaining a current image of the target area;

[0064] S170, determining the number of scales on the surface of the mica pulp based on the current image;

[0065] S180, determining a second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp;

[0066] S190 , determining a final concentration range of the mica pulp in combination with the first concentration range and the second concentration range, so as to generate a quality control strategy according to the final concentration range.

[0067] A first control instruction is issued to a vibrator located at the bottom of the mica pulp pool, causing the mica pulp to vibrate. In this embodiment, the first control instruction is a control instruction to activate the vibrator at the bottom of the mica pulp pool. After the mica pulp vibrates, an image of a target area of ​​the mica pulp is acquired. Image recognition technology is used to determine vibration ripple characteristics of the target area image. In this embodiment, the target area image is an image of the mica pulp surface after the vibrator is activated. Image recognition refers to the use of computers and artificial intelligence technologies to analyze and understand images to automatically identify and classify information such as objects, scenes, and patterns within the image. By acquiring the target area image of the mica pulp and using image recognition technology, vibration ripple characteristics of the target area image are determined. In this embodiment, the vibration ripple characteristics include the number of vibration ripples and the amplitude of the vibration ripples.

[0068] Based on the vibration ripple characteristics of the target area image, a first concentration range of the mica pulp is determined, and it is determined whether the first concentration range is within a preset concentration range. In this embodiment, the first concentration range of the mica pulp is determined by analyzing the vibration ripple characteristics of the target area image of the mica pulp. Mica pulp concentration refers to the concentration of solid matter in the mica pulp in water during beating. Mica pulp concentration is a relative term, representing the ratio of mica to other fibers per 100 ml of pulp. The first concentration range of the mica pulp is determined based on the vibration ripple characteristics of the target area image. Specifically, the first concentration of the mica pulp is determined by comparing the number of vibration ripples and amplitudes in the target area with the number of vibration ripples at a preset concentration. In this embodiment, the preset concentration can be determined based on time. The number of vibration ripples at the preset concentration is the number of vibration ripples corresponding to each preset concentration stored in a database. For example, if the number of vibration ripples in the target area is 35 and the preset concentration in the database is 2%, the corresponding number of vibration ripples is 35. Therefore, the first concentration of the mica pulp is determined to be 2%. Similarly, the vibration ripple amplitude is compared with the vibration ripple amplitude of the preset concentration to obtain the second concentration of the mica pulp. In this embodiment, the vibration ripple amplitude of the preset concentration is the vibration ripple amplitude corresponding to each preset concentration stored in the database. For example, if the vibration ripple amplitude of the target area is 10 cm, and the preset concentration in the database is 3%, the corresponding vibration ripple amplitude is 10 cm. Therefore, the second concentration of the mica pulp is determined to be 3%. The first concentration range of the mica pulp is determined by combining the first and second concentrations of the mica pulp. For example, if the first concentration of the mica pulp is 2% and the second concentration of the mica pulp is 3%, the first concentration range of the mica pulp is determined to be 2%-3%. After determining the first concentration range of the mica pulp, it is determined whether the first concentration range is within the preset concentration range. In this embodiment, the preset concentration range can be determined based on actual conditions. For example, if the preset concentration range is 2%-5% and the first concentration range of the mica pulp is 2%-3%, it can be determined that the first concentration range is within the preset concentration range.

[0069] If the first concentration range is not within the preset concentration range, a warning command is issued. Mica pulp concentration refers to the concentration of mica pulp solid matter in water during beating. Mica pulp concentration is a relative concept, representing the ratio of mica to other fibers per 100 ml of pulp. Specifically, the obtained first concentration range of mica pulp is compared with the preset concentration range to determine the warning level for the first concentration range. For example, if the first concentration range of mica pulp is 1%-3% and the preset concentration range is 5%-7%, the first concentration range is not within the preset concentration range, the average value of the first concentration range is 2%, the average value of the preset concentration range is 6%, and the preset concentration range is three times the first concentration range. The warning level can be primary, intermediate, or high. If the preset concentration range is three times the first concentration range, it is a primary warning level, and a primary warning command for the mica pulp concentration is issued. The issuance of the primary warning command can be broadcast to notify staff at the production site as a warning prompt.

[0070] After a warning is issued after determining that the first concentration range is outside the preset range, staff will adjust the first concentration range of the mica pulp. First, the concentration of the first concentration range is determined to determine whether it is greater than or less than the preset range. If the first concentration range exceeds the preset range, this indicates that the concentration of mica pulp solids in water is too high, and the pulp concentration needs to be reduced by increasing the amount of water. Specifically, the total water supply required for dilution is determined based on the amount of pulp in the mixing tank, and the water supply rate is set. The pulp is stirred and water supply is initiated. The current water supply rate is determined based on the current water consumption. This method is divided into two stages: the first stage is conducted at a lower water supply rate. When the water consumption reaches 40%-70% of the total, the second stage is switched to a higher water supply rate to ensure uniform mixing of the pulp and water. The second stage determines whether the first concentration range is within the preset range. Adjustment of the mica pulp concentration ceases until the first concentration range of the mica pulp is within the preset range.

[0071] When the first concentration range is less than the preset concentration range, it indicates that the concentration of mica pulp solid matter in water in the first concentration range is too low, and the proportion of mica and other fibers needs to be increased appropriately to increase the pulp concentration. In addition, the concentration of mica pulp can be adjusted by mechanical pressing, which is a commonly used pulp concentration adjustment method. In this method, mechanical pressing equipment can be used to press the pulp to gradually drain the water therein, thereby adjusting the pulp concentration. After the water is squeezed out using mechanical pressing equipment, it is determined whether the first concentration range is within the preset concentration range, and the concentration of the mica pulp is stopped until the first concentration range of the mica pulp is within the preset concentration range. ‌‌

[0072] If the first concentration range is within the preset concentration range, after the mica pulp is allowed to stand for a preset time period, a second control instruction is issued to control the light source to emit incident light to the target area. In this embodiment, the second control instruction refers to an instruction to control the light source to emit incident light to the target area. The preset time period refers to the time period when the vibration ripples on the surface of the mica pulp disappear. That is, when the first concentration range is within the preset concentration range, after the mica pulp is allowed to stand for a preset time period, when the vibration ripples on the surface of the mica pulp disappear, the light source emits incident light to the target area to reduce the impact of the vibration ripples of the mica pulp on the incident light. In this embodiment, the light source can be an artificial light source, such as a fluorescent lamp and a laser, and a current image of the target area of ​​the mica pulp is obtained.

[0073] The number of scales on the mica pulp surface is determined using a current image. Specifically, first, grayscale processing is performed on the current image to obtain a grayscale image, wherein the current image includes multiple reflected light points obtained based on incident light. A third pixel of the grayscale image and a fourth pixel of the background image of the mica pulp surface are then obtained. In this embodiment, the third pixel refers to a pixel of a light source reflection point formed by controlling a light source to emit incident light onto the mica pulp surface in the grayscale image, and the fourth pixel refers to a pixel in the background of the mica pulp surface image. The third and fourth pixels can be obtained through image recognition. Next, the pixels of the current image are obtained and color characteristics of all pixels in the current image are identified. For example, the color characteristics of the pixels of the light source reflection point are black, while the color characteristics of the pixels in the background of the mica pulp surface image are white. The pixels of the reflected light points are then determined based on the color characteristics. Using a preset edge detection algorithm, the edges of the pixels of the reflected light points are detected. Based on the edges of the reflected light points, the number of reflected light points is determined. The number of reflected light points is used to indicate the number of scales on the mica pulp surface.

[0074] According to the number of scales on the surface of the mica pulp, the second concentration range of the mica pulp is determined. Specifically, the number of scales on the surface of the mica pulp is retrieved in a preset database to obtain the mica pulp concentration corresponding to the number of scales on the surface of the mica pulp, and the retrieved concentration of the mica pulp is used as the second concentration range. In this embodiment, the second concentration range is the concentration range of the mica pulp obtained by retrieving the concentration of the mica pulp corresponding to the number of scales on the surface of the mica pulp in the database. For example, it is determined that the number of scales on the surface of the mica pulp is 530, and the mica pulp concentration corresponding to the number of scales on the surface of the mica pulp 530 in the preset database is 3%~5%, so the mica pulp concentration is determined to be 3%~5%.

[0075] The final concentration range of the mica pulp is determined based on the first and second concentration ranges of the mica pulp, and a quality control strategy is generated based on the final concentration range. Specifically, when the second concentration range of the mica pulp falls within the first concentration range, the second concentration range is used as the final concentration range of the mica pulp. When the first and second concentration ranges of the mica pulp overlap, the overlapping range of the first and second concentration ranges is used as the final concentration range of the mica pulp. When the first and second concentration ranges of the mica pulp are adjacent, the range obtained by adding the first and second concentration ranges is used as the final concentration range of the mica pulp. An alarm is triggered for the mica pulp concentration, notifying staff to retest the first and second concentration ranges of the mica pulp. Generating a quality control strategy based on the final concentration range specifically means that when the final concentration range of the mica pulp is too high, indicating that the solid content of the mica pulp is greater than the liquid content, a quality control strategy is generated to add liquid to the mica pulp to reduce the final concentration range of the mica pulp. Similarly, when the final concentration range of the mica pulp is low, it indicates that the solid matter content in the mica pulp is less than the liquid content in the mica pulp. At this time, solid matter needs to be added to the mica pulp according to the generated quality control strategy to increase the final concentration range of the mica pulp.

[0076] The above technical solution uses a vibrator at the bottom of the mica pulp tank to vibrate the mica pulp, capturing a target area on the mica pulp surface. The vibration ripple characteristics of the target area image are then determined using image recognition. This allows for a more precise quality control of the mica pulp concentration. Based on the vibration ripple characteristics, a first concentration range of the mica pulp is determined. Obtaining the first concentration range facilitates more accurate concentration analysis of the mica pulp, enhancing mica pulp quality control. If the first concentration range falls outside the preset concentration range, an early warning is issued, alerting personnel and helping to promptly identify potential risks associated with the mica pulp. By determining the number of scales on the mica pulp surface, a second concentration range of the mica pulp is determined, and a final concentration range is determined. A quality control strategy is generated based on the final concentration range, enabling precise control of the mica pulp concentration, effectively ensuring accurate concentration during production.

[0077] In one implementation of this embodiment, performing image recognition on the target area image to obtain the vibration ripple features of the target area includes:

[0078] S210, dividing the target area image into a plurality of unit grid images, and preprocessing each unit grid image to obtain a corresponding grayscale unit grid image;

[0079] S220 , performing image recognition on each grayscale unit grid image to obtain the number of vibration ripples and the amplitude of the vibration ripples.

[0080] The target area image is divided into multiple unit grid images, and each unit grid image is preprocessed to obtain a corresponding grayscale unit grid image. That is, the target area image of the mica pulp surface is divided into multiple unit grid images. In this embodiment, the target area image refers to the area image without contaminated area on the mica pulp surface. The image without contaminated area is divided to obtain multiple unit grid images, and each unit grid image is preprocessed to obtain multiple grayscale unit grid images.

[0081] Image recognition uses computers and artificial intelligence to analyze and understand images, automatically identifying and classifying objects, scenes, patterns, and other information within them. A preset image recognition algorithm can be used to perform image recognition on each grayscale unit grid image, determining the number and amplitude of vibration ripples within each grayscale unit grid image.

[0082] By dividing the target area image into multiple unit grid images and preprocessing each unit grid image to obtain the corresponding grayscale unit grid image, the target area image can be made clearer. By image recognition of each grayscale unit grid image, the number of vibration ripples and the vibration ripple amplitude can be obtained, which can make the analysis of the target area image more accurate and ensure the accuracy of the obtained vibration ripple number and vibration ripple amplitude data.

[0083] In one implementation of this embodiment, after preprocessing each unit grid image to obtain a corresponding grayscale unit grid image, the method further includes:

[0084] S310, for any grayscale unit grid image, performing denoising processing on the grayscale unit grid image;

[0085] S320, performing feature extraction on the grayscale unit grid image after denoising to obtain target impurity edge features;

[0086] S330, obtaining a first pixel point of a target impurity edge feature and a second pixel point of a mica flake edge feature;

[0087] S340, calculating the Euclidean distance and color contrast between the first pixel and the second pixel, and determining whether the target area is a contaminated area based on the Euclidean distance and color contrast between the first pixel and the second pixel;

[0088] S350: If the Euclidean distance and the color contrast are both greater than a preset threshold, the target area is determined to be a contaminated area and an alarm is issued;

[0089] S360: If the Euclidean distance and the color contrast are both less than or equal to the preset threshold, it is determined that the target area is not a contaminated area, and image recognition is performed on each grayscale unit grid image to obtain the number of vibration ripples and the amplitude of the vibration ripples.

[0090] Figure 2 A schematic diagram of a process for detecting a contaminated area in a target area according to an embodiment of the present application is shown schematically. Figure 2 As shown, after issuing a first control instruction to activate the vibrator, a target area on the surface of the mica pulp is obtained and a determination is made as to whether there is a contaminated area within the target area. If there is a contaminated area within the target area, the contaminated area is identified and a warning message is issued. After removing the contaminated area and confirming that there is no contaminated area within the target area, the instruction for identifying vibration ripples is executed. If there is no contaminated area within the target area, the instruction for identifying vibration ripples is directly executed.

[0091] For any grayscale unit grid image, the grayscale unit grid image is denoised. In this embodiment, the grayscale unit grid image is denoised mainly through a pre-built deep convolutional neural network model, and the pre-built model is trained with a large amount of image data to achieve efficient image denoising.

[0092] After denoising the grayscale unit grid image, feature extraction is performed on the denoised grayscale unit grid image to obtain the target impurity edge features. In this embodiment, the target impurity edge features can be the edge features of suspended matter in water during the production process, such as the edge features of materials such as plastic, wood chips, bark, and weeds. Using image recognition technology, the image gradient of the target impurity is obtained. The image gradient refers to a vector obtained by treating the image as a two-dimensional discrete function and calculating the rate of change of each pixel in the image in the X-axis and Y-axis directions. This vector is used to represent the edge information of the image at that point and is commonly used in tasks such as edge detection, feature extraction, and image enhancement. The target impurity edge features are determined by the image gradient of the target impurity.

[0093] The first pixel point of the target impurity edge feature and the second pixel point of the mica flake edge feature are obtained. In this embodiment, the first pixel point refers to the pixel point of the edge feature of the impurity on the surface of the mica pulp, and the second pixel point is the pixel point of the edge feature of the mica flake on the surface of the mica pulp. The first pixel point of the target impurity edge feature and the second pixel point of the mica flake edge feature can be obtained through image recognition technology.

[0094] After obtaining the first pixel point of the target impurity edge feature and the second pixel point of the mica scale edge feature, the Euclidean distance and color contrast between the first pixel point and the second pixel point are calculated, wherein the Euclidean distance is used to calculate the true distance between two points in m-dimensional space. In this embodiment, the Euclidean distance between the first pixel point and the second pixel point can be calculated by the coordinates of the first pixel point and the second pixel point. In two-dimensional space, the calculation formula of the Euclidean distance is:

[0095] ‌

[0096] Wherein, x1 and y1 represent the horizontal coordinate and vertical coordinate of the first pixel point in the two-dimensional space, and x2 and y2 represent the horizontal coordinate and vertical coordinate of the second pixel point in the two-dimensional space.

[0097] Calculating color contrast refers to measuring the different brightness levels between the brightest white and the darkest black in the light and dark areas of an image. In this embodiment, calculating the color contrast between the first pixel and the second pixel refers to calculating the color contrast between the first pixel of the target impurity edge feature and the second pixel of the mica scale edge feature. For example, the color of the first pixel of the target impurity edge feature is black, and the second pixel of the mica scale edge feature is white. By calculating the color contrast between the first pixel and the second pixel, the mica scale and the target impurity can be effectively distinguished, which facilitates the accurate detection of the quality of the mica pulp concentration.

[0098] Based on the Euclidean distance and color contrast between the first pixel and the second pixel, determine whether the target area is a contaminated area. In this embodiment, the contaminated area refers to the area where impurities are located in the mica pulp. That is, by calculating the Euclidean distance and color contrast between the first pixel and the second pixel, the obtained Euclidean distance and color contrast are compared with the preset threshold. If both the Euclidean distance and the color contrast are greater than the preset threshold, the target area is determined to be a contaminated area. The preset threshold can be determined according to production requirements. For example, the calculated Euclidean distance between the first pixel and the second pixel is 5, and the preset Euclidean distance threshold is 10, indicating that the Euclidean distance between the first pixel and the second pixel is too large, and there are impurities in the current image. The calculated color contrast between the first pixel and the second pixel is 120:1, and the preset color contrast threshold is 50:1, indicating that the current color contrast between the first pixel and the second pixel exceeds the preset threshold. When both the Euclidean distance and the color contrast are greater than the preset threshold, the target area is determined to be a contaminated area.

[0099] After confirming the presence of a contaminated area, an alarm is issued. That is, when a contaminated area exists in the mica pulp, an early warning message is issued to the staff. The corresponding early warning message can be sent via mobile phone text message using the staff's reserved mobile phone number as the first emergency contact. After confirming that there is no contaminated area in the target area, image recognition is performed on each grayscale unit grid image to obtain the number and amplitude of vibration ripples.

[0100] By detecting whether there is a contaminated area in the target area, the quality detection of the mica pulp concentration can be further monitored. When there is no contaminated area in the target area of ​​the mica pulp, the step of performing image recognition on the grayscale unit grid image to obtain the number and amplitude of vibration ripples can make the number and amplitude of vibration ripples obtained more accurate, thereby ensuring the accuracy of the mica pulp concentration detection.

[0101] In one implementation of this embodiment, determining the first concentration range of the mica pulp according to the vibration ripple characteristics of the target area includes:

[0102] S410, obtaining the number of vibration ripples and the amplitude of vibration ripples based on the vibration ripple characteristics of the target area;

[0103] S420, comparing the number of vibration ripples with the number of vibration ripples of a preset concentration to obtain a first concentration of the mica pulp;

[0104] S430, comparing the vibration ripple amplitude with the vibration ripple amplitude of a preset concentration to obtain a second concentration of the mica pulp;

[0105] S440: Determine a first concentration range of the mica pulp based on the first concentration of the mica pulp and the second concentration of the mica pulp.

[0106] After obtaining the vibration ripple characteristics of the target area, the number of vibration ripples and the amplitude of the vibration ripples are determined. Specifically, the vibration ripple characteristics are obtained through image recognition technology, and the number of vibration ripples and the amplitude of the vibration ripples on the surface of the mica pulp are determined.

[0107] The number of vibration ripples is then compared with the number of vibration ripples for a preset concentration to obtain a first concentration of the mica pulp. The mica pulp concentration refers to the concentration of solid matter in the mica pulp in water during beating. The definition of mica pulp concentration is a relative concept, representing the ratio of mica to other fibers in 100 ml of pulp. For example, a mica pulp concentration of 2% indicates that mica powder and other papermaking fibers account for 2% of the mica pulp in 100 ml of pulp. In this embodiment, the preset concentration can be determined based on actual conditions. The number of vibration ripples for the preset concentration is the number of vibration ripples corresponding to each preset concentration stored in the database. For example, a preset concentration of 1% in the database corresponds to 20 vibration ripples, while a preset concentration of 2% corresponds to 35 vibration ripples. The number of vibration ripples is compared with the number of vibration ripples for the preset concentration to obtain the first concentration of the mica pulp. For example, if the number of vibration ripples in the target area is 35, and the preset concentration range corresponding to a number of vibration ripples of 35 in the database is 3%, then the first concentration of the mica pulp is determined to be 3%.

[0108] The vibration ripple amplitude is compared with the vibration ripple amplitude of a preset concentration to obtain the second concentration of the mica pulp. Specifically, the obtained vibration ripple amplitude is compared with the vibration ripple amplitude of a preset concentration in a database. In this embodiment, the preset concentration can be determined based on actual conditions. The vibration ripple amplitude of the preset concentration is the vibration ripple amplitude corresponding to each preset concentration stored in the database. For example, a vibration ripple amplitude of 1% corresponding to a preset concentration in the database is 10, and a vibration ripple amplitude of 2% corresponding to a preset concentration is 30. The number of vibration ripples is compared with the number of vibration ripples of the preset concentration to obtain the first concentration of the mica pulp. For example, if the vibration ripple amplitude in the target area is 10 cm, and the concentration in the preset concentration range in the database is 1%, the corresponding vibration ripple amplitude is 10 cm. Therefore, the second concentration of the mica pulp is determined to be 1%.

[0109] The first concentration range of the mica pulp is determined by combining the first concentration of the mica pulp and the second concentration of the mica pulp. In this embodiment, the first concentration range is the mica pulp concentration range determined by the first concentration and the second concentration of the mica pulp.

[0110] Combined with the first concentration and the second concentration of the mica pulp, the first concentration range of the mica pulp is determined. For example, the first concentration of the mica pulp is 3%, and the second concentration of the mica pulp is 1%. The first concentration range of the mica pulp is determined to be 1% to 3%.

[0111] By obtaining the vibration ripple characteristics of the target area, comparing the number of vibration ripples with the number of vibration ripples in a preset concentration range, the first concentration of the mica pulp is obtained, and comparing the vibration ripple amplitude with the vibration ripple amplitude in the preset concentration range, the second concentration of the mica pulp is obtained. Combining the first concentration of the mica pulp and the second concentration of the mica pulp to determine the first concentration range of the mica pulp, the concentration of the mica pulp can be obtained more accurately, the quality detection effect of the mica pulp is improved, and the production quality of the mica paper is guaranteed.

[0112] In one implementation of this embodiment, if the first concentration range is not within the preset concentration range, issuing a warning instruction of the mica pulp concentration includes:

[0113] S510, comparing the first concentration range with a preset concentration range to determine a warning level for the first concentration range;

[0114] S520. Based on the warning level, issue a warning instruction for mica pulp concentration.

[0115] When it is determined that the first concentration range is not within the preset concentration range, a warning instruction for the mica pulp concentration is issued. The preset concentration range can be determined based on actual conditions. Specifically, the first concentration range is compared with the preset concentration range to determine the warning level for the first concentration range. That is, different warning levels are generated for mica pulp according to different preset concentration ranges. In this embodiment, the warning levels can be divided into primary warning level, intermediate warning level, and advanced warning level, and different warning levels correspond to different warning concentration ranges. For example, when the concentration exceeds the preset concentration range by ±3 times, the primary warning level is set. The primary warning level can be broadcast to notify the staff at the production site that the current first concentration range is not within the preset concentration range and exceeds ±3 times the preset concentration range. When the concentration exceeds the preset concentration range by ±6 times, the intermediate warning level is set. The intermediate warning level can be notified by phone and broadcast to the staff who have reserved mobile phone numbers and the staff at the production site, and the staff's reserved mobile phone number is called to remind the staff that the current first concentration range of mica pulp is not within the preset concentration range and exceeds ±6 times the preset concentration range. When the concentration exceeds the preset range by ±9 times, it is a high-level warning level. The high-level warning level can be notified to the staff through broadcasting, calling the staff with the reserved mobile phone number, or sending text messages that the current mica pulp first concentration range is not within the preset concentration range and exceeds the preset concentration range by ±9 times.

[0116] Based on the warning level, a warning instruction for mica pulp concentration is issued. That is, if the first concentration range of mica pulp obtained by determination is within the corresponding warning level, different warning level notifications are issued through different means. For example, the first concentration range of mica pulp is 1%-3%, the preset concentration range is 5%-7%, the first concentration range is not within the preset concentration range, the average value of the first concentration range is 2%, the average value of the preset concentration range is 6%, and the preset concentration range is three times the first concentration range. The warning levels can be primary, intermediate, and high. The preset concentration range is three times the first concentration range, which is the primary warning level. Therefore, a primary warning instruction for mica pulp concentration is issued. The issuance of the primary warning instruction can be broadcast to notify the staff at the production site as a warning prompt.

[0117] By comparing the first concentration range with the preset concentration range, determining the warning level of the first concentration range, and issuing a warning instruction for the mica pulp concentration, the risk hazards of mica pulp in quality control can be discovered in time, the inaccurate concentration of mica pulp can be prevented, and relevant staff can be notified in time to deal with the concentration of mica pulp, thereby ensuring the accuracy of the concentration of mica pulp during the production process.

[0118] In one implementation manner of this embodiment, determining the number of scales on the surface of the mica pulp using the current image includes:

[0119] S610, performing grayscale processing on the current image to obtain a grayscale image, wherein the current image includes a plurality of reflected light points obtained based on incident light;

[0120] S620, obtaining a third pixel point of the grayscale image and a fourth pixel point of the background image of the mica pulp surface; obtaining pixel points of the current image, and identifying color features of all pixel points of the current image;

[0121] S630, determining the pixel point of the reflected light point according to the color feature;

[0122] S640. Determine the edge of the reflected light spot using a preset edge detection algorithm based on the pixel points of the reflected light spot, and obtain the number of reflected light spots based on the edge of the reflected light spot, wherein the number of reflected light spots is used to characterize the number of scales on the surface of the mica pulp.

[0123] The number of scales on the mica pulp surface is determined using the current mica pulp surface image. Specifically, the current image is first grayscale processed to obtain a grayscale image. In this embodiment, the grayscale image is a grayscale image of the mica pulp surface. Grayscale processing of the current image can be performed using an average method, which takes the arithmetic average of the pixel values ​​in the original image; the average result is the grayscale value of that pixel. A grayscale image includes multiple reflected light points based on incident light. In other words, the grayscale image contains multiple reflected light points. These reflected light points are the reflections of incident light emitted by a light source onto a target area by the mica scales on the mica pulp surface. Grayscale processing of the current image to obtain a grayscale image facilitates a clearer analysis of the mica pulp surface image.

[0124] A third pixel of the grayscale image and a fourth pixel of the background image of the mica pulp surface are obtained. In this embodiment, the third pixel of the grayscale image refers to a pixel of the grayscale image of mica flakes on the mica pulp surface, and the fourth pixel refers to a pixel in the background image of the mica pulp surface. The third pixel of the grayscale image and the fourth pixel of the background image of the mica pulp surface are obtained, and based on the pixels of the current image, the color features of all pixels of the current image are identified. In other words, the color features of all pixels of the current image are determined by analyzing the obtained color features of the third and fourth pixels. For example, the color feature of the third pixel of the grayscale image is obtained to be black, and the color feature of the fourth pixel of the background image of the mica pulp surface is obtained to be white. The color features of the pixels of the current image are identified using image recognition technology. Image recognition refers to the use of computers and artificial intelligence technologies to analyze and understand images to automatically identify and classify information such as objects, scenes, and patterns in the images.

[0125] The color features obtained from the current image are compared with the color features of the third pixel point of the grayscale image and the color features of the fourth pixel point of the background image of the mica pulp surface, and the pixel point of the reflected light point in the current image is determined based on the color features. For example, the color feature of the reflected light point pixel point in the current image is black, and the color feature of the fourth pixel point of the background image of the mica pulp surface is white. By determining the areas in the current image where all color features are black, these areas are used as the locations of the reflected light points.

[0126] Based on the pixels of the reflected light spot, a preset edge detection algorithm is used to determine the edge of the reflected light spot. Specifically, the position of the reflected light spot can be determined based on color features, and a preset edge detection algorithm can be used. In this embodiment, the preset edge detection algorithm can be a Sobel operator, which is primarily used for edge detection. The edge of the reflected light spot is determined using the Sobel operator. For example, if the edge of the reflected light spot has an irregular elliptical shape, the position of the irregular elliptical edge in the current image is determined, thereby determining the position of the reflected light spot in the current image.

[0127] According to the edges of the reflected light spots, the number of reflected light spots is obtained, wherein the number of reflected light spots is used to characterize the number of scales on the surface of the mica pulp. That is, by determining the characteristics of the edges of the reflected light spots, according to the characteristics of the edges of the reflected light spots, the positions of the reflected light spots are determined in the current image, and the number of positions of the reflected light spots is counted, thereby determining the number of reflected light points.

[0128] By identifying the color features in the current image and using image recognition technology to obtain the number of reflected light spots on the surface of the mica pulp, the number of mica flakes on the surface of the mica pulp can be accurately obtained, which helps to determine the accuracy of the concentration of the mica pulp through the number of mica flakes and ensure the accuracy of the mica pulp quality control.

[0129] In one implementation of this embodiment, determining the second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp includes:

[0130] S710. Retrieve the number of scales on the surface of the mica pulp from a preset database to obtain a second concentration range of the mica pulp corresponding to the number of scales on the surface of the mica pulp.

[0131] The second concentration range of the mica pulp is determined by the number of scales on the surface of the mica pulp obtained previously. The mica pulp concentration refers to the concentration of the solid matter of the mica pulp in water during beating. The mica pulp concentration is a relative concept, which is the ratio of mica to other fibers in every 100 ml of pulp. For example, a mica pulp concentration of 2% means that the proportion of mica powder and other papermaking fibers in every 100 ml of pulp is 2%. In this embodiment, the second concentration range is obtained by searching a preset database for the concentration range of the mica pulp corresponding to the number of scales on the mica pulp surface. For example, the number of scales on the mica pulp surface determined by image recognition technology is 200. The number of scales on the mica pulp surface is retrieved from the database, and the concentration of the mica pulp corresponding to the number of scales is 4%-5%. Therefore, it can be determined that when the number of mica scales on the mica pulp surface is 200, the concentration of the mica pulp is 4%-5%. The concentration of mica pulp can be determined by the number of mica flakes on the surface of the mica pulp. The concentration of the liquid can be determined by utilizing the Fresnel reflection law. The Fresnel reflection law determines the concentration of the liquid by measuring the reflection intensity of light. In this embodiment, the concentration of the mica pulp is determined by the reflectivity of the mica flakes.

[0132] By searching the concentration of mica pulp based on mica flakes in the database, the concentration of mica pulp can be effectively determined based on the number of scales on the surface of the mica pulp. This not only achieves accurate identification of the number of mica flakes and improves the accuracy of identification, but also reduces the inaccurate detection of mica pulp concentration, which helps to improve the safety and reliability of mica pulp quality control.

[0133] In one implementation of this embodiment, determining the final concentration range of the mica pulp in combination with the first concentration range and the second concentration range of the mica pulp includes:

[0134] S810: When the second concentration range of the mica pulp is within the first concentration range, setting the second concentration range as the final concentration range of the mica pulp;

[0135] S820: When the first concentration range and the second concentration range of the mica pulp overlap, use the overlapping range of the first concentration range and the second concentration range as the final concentration range of the mica pulp;

[0136] S830: When the first concentration range and the second concentration range of the mica pulp are adjacent to each other, a range obtained by adding the first concentration range and the second concentration range is used as the final concentration range of the mica pulp.

[0137] A first concentration range and a second concentration range of mica pulp are obtained, and a final concentration range of mica pulp is determined. In this embodiment, the first concentration range of mica pulp refers to the mica pulp concentration range obtained by determining the first concentration and the second concentration of mica pulp, and the second concentration range is the mica pulp concentration range obtained by searching a database for the concentration of mica pulp corresponding to the number of scales on the surface of the mica pulp. The final concentration range refers to the mica pulp concentration range finally determined by the first concentration range and the second concentration range of mica pulp.

[0138] In combination with the first concentration range and the second concentration range of the mica pulp, the final concentration range of the mica pulp is determined. Specifically, when the second concentration range of the mica pulp is included in the first concentration range, the second concentration range is used as the final concentration range of the mica pulp. That is, when the first concentration range of the mica pulp is greater than the second concentration range of the mica pulp, the second concentration range is used as the final concentration range of the mica pulp. For example, when the first concentration range of the mica pulp is [1%, 5%] and the second concentration range of the mica pulp is [2%, 3%], the second concentration range of the mica pulp is within the first concentration range of the mica pulp. Therefore, the second concentration range of the mica pulp is [2%, 3%] as the final concentration range of the mica pulp, and the concentration range of the mica pulp is narrowed, making the concentration of the mica pulp more accurate.

[0139] When the first concentration range and the second concentration range of mica pulp overlap, the overlapping range of the first concentration range and the second concentration range is used as the final concentration range of the mica pulp. Specifically, when the first concentration range and the second concentration range of mica pulp overlap, the overlapping part of the first concentration range and the second concentration range is used as the final concentration range of the mica pulp, so that the concentration range of the mica pulp is reduced and the concentration of the mica pulp is more accurate. For example, the first concentration range of mica pulp is [1%, 5%], the second concentration range of mica pulp is [2%, 7%], and the overlapping concentration range of the first concentration range and the second concentration range of mica pulp is [2%, 5%]. Therefore, the overlapping concentration range of the mica pulp is [2%, 5%], which is used as the final concentration range of the mica pulp.

[0140] When the first concentration range and the second concentration range of the mica pulp are adjacent, the range obtained by adding the first concentration range and the second concentration range is used as the final concentration range of the mica pulp. That is, when the first concentration range of the mica pulp is [1%, 5%], the second concentration range of the mica pulp is [6%, 7%]. The range obtained by adding the first concentration range and the second concentration range of the mica pulp is used as the final concentration range of the mica pulp. Therefore, the final concentration range of the mica pulp is [1%, 7%]. However, after combining the first concentration range and the second concentration range of the mica pulp, the final concentration range of the mica pulp becomes larger, resulting in the final concentration range of the mica pulp becoming inaccurate. Therefore, when the first concentration range and the second concentration range of the mica pulp are adjacent, it is an abnormal situation of the mica pulp concentration range, indicating that the first concentration range or the second concentration range of the mica pulp is inaccurate, resulting in the final concentration range of the mica pulp becoming larger. When the first concentration range and the second concentration range of the mica pulp are adjacent to each other, an alarm of concentration detection error is issued, and the staff re-tests the first concentration range and the second concentration range of the mica pulp to determine the latest first concentration range and the second concentration range, and determines the final concentration range of the mica pulp in combination with the latest first concentration range and the second concentration range.

[0141] By combining the first concentration range and the second concentration range of the mica pulp to determine the final concentration range of the mica pulp, the error in the detection of the mica pulp concentration can be reduced, the detection of the mica pulp can be made more accurate, the accuracy of the mica pulp concentration in the mica paper production process can be guaranteed, and the quality of the produced mica paper can be guaranteed.

[0142] An embodiment of the present application further provides a machine-readable storage medium having stored thereon instructions for causing a machine to execute the above-mentioned mica pulp quality control method based on image recognition.

[0143] The present application also provides a mica pulp quality control system based on image recognition, including:

[0144] a memory configured to store instructions; and

[0145] The processor is configured to call instructions from the memory and implement the above-mentioned mica pulp quality control method based on image recognition when executing the instructions.

[0146] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0150] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0151] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0152] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0153] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0154] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A mica pulp quality control method based on image recognition, characterized in that: include: issuing a first control instruction to control the operation of a vibrator located at the bottom of the mica pulp tank; Acquiring a target area image of mica pulp, performing image recognition on the target area image, and obtaining vibration ripple features of the target area image, wherein the vibration ripple features include the number of vibration ripples and the amplitude of the vibration ripples; determining a first concentration range of mica pulp according to the vibration ripple characteristics of the target area image, and judging whether the first concentration range is within a preset concentration range; If the first concentration range is not within the preset concentration range, issuing an early warning instruction; If the first concentration range is within a preset concentration range, after the mica pulp is allowed to stand for a preset period of time, a second control instruction is issued to control the light source to emit incident light toward a target area; Acquire a current image of the target area; determining the number of scales on the surface of the mica pulp using the current image; determining a second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp; The first concentration range and the second concentration range of the mica pulp are combined to determine a final concentration range of the mica pulp, so as to generate a quality control strategy according to the final concentration range.

2. The method according to claim 1, characterized in that The performing image recognition on the target area image to obtain the vibration ripple characteristics of the target area includes: Dividing the target area image into a plurality of unit grid images, and preprocessing each of the unit grid images to obtain a corresponding grayscale unit grid image; Image recognition is performed on each of the grayscale unit grid images to obtain the number of vibration ripples and the amplitude of the vibration ripples.

3. The method according to claim 2, characterized in that After preprocessing each of the unit grid images to obtain a corresponding grayscale unit grid image, the method further includes: For any one of the grayscale unit grid images, performing denoising processing on the grayscale unit grid image; Performing feature extraction on the grayscale unit grid image after denoising to obtain target impurity edge features; Acquire a first pixel point of the target impurity edge feature and a second pixel point of the mica flake edge feature; Calculating the Euclidean distance and color contrast between the first pixel and the second pixel, and determining whether the target area is a contaminated area based on the Euclidean distance and color contrast between the first pixel and the second pixel; If the Euclidean distance and the color contrast are both greater than a preset threshold, the target area is determined to be a polluted area and an alarm message is issued; If the Euclidean distance and the color contrast are both less than or equal to a preset threshold, it is determined that the target area is not a contaminated area, and the step of performing image recognition on each grayscale unit grid image to obtain the number of vibration ripples and the amplitude of the vibration ripples is performed.

4. The method according to claim 1, wherein Determining a first concentration range of mica pulp according to the vibration ripple characteristics of the target area includes: Obtaining the number of vibration ripples and the amplitude of vibration ripples according to the vibration ripple characteristics of the target area; Comparing the number of vibration ripples with the number of vibration ripples of a preset concentration to obtain a first concentration of mica pulp; Comparing the vibration ripple amplitude with the vibration ripple amplitude of a preset concentration to obtain a second concentration of the mica pulp; The first concentration range of the mica pulp is determined by combining the first concentration of the mica pulp and the second concentration of the mica pulp.

5. The method according to claim 1, wherein If the first concentration range is not within a preset concentration range, issuing a warning instruction of the mica pulp concentration includes: comparing the first concentration range with a preset concentration range to determine a warning level for the first concentration range; Based on the warning level, a warning instruction for mica pulp concentration is issued.

6. The method according to claim 1, characterized in that Determining the number of scales on the surface of the mica pulp using the current image includes: Performing grayscale processing on the current image to obtain a grayscale image, wherein the current image includes a plurality of reflected light points obtained based on the incident light; Acquire a third pixel point of the grayscale image and a fourth pixel point of the background image of the mica pulp surface; acquire pixel points of the current image, and identify color features of all the pixel points of the current image; Determine the pixel point of the reflected light point according to the color feature; According to the pixel points of the reflected light spots, a preset edge detection algorithm is used to determine the edges of the reflected light spots, and based on the edges of the reflected light spots, the number of the reflected light spots is obtained, wherein the number of the reflected light points is used to characterize the number of scales on the surface of the mica pulp.

7. The method according to claim 6, characterized in that Determining the second concentration range of the mica pulp according to the number of scales on the surface of the mica pulp includes: The number of scales on the surface of the mica pulp is retrieved in a preset database to obtain a second concentration range of the mica pulp corresponding to the number of scales on the surface of the mica pulp.

8. The method according to claim 7, characterized in that The step of determining a final concentration range of the mica pulp by combining the first concentration range and the second concentration range of the mica pulp comprises: When the second concentration range of the mica pulp is included in the first concentration range, the second concentration range is set as the final concentration range of the mica pulp; When the first concentration range and the second concentration range of the mica pulp overlap, the overlapping range of the first concentration range and the second concentration range is used as the final concentration range of the mica pulp; When the first concentration range and the second concentration range of the mica pulp are adjacent to each other, the range obtained by adding the first concentration range and the second concentration range is used as the final concentration range of the mica pulp.

9. A machine-readable storage medium, characterized in that The machine-readable storage medium stores instructions for enabling a machine to execute the mica pulp quality control method based on image recognition according to any one of claims 1 to 8.

10. A mica pulp quality control system based on image recognition, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and implement the mica pulp quality control method based on image recognition according to any one of claims 1 to 8 when executing the instructions.

Images