Potassium feldspar ceramic sintering whiteness detection system and method

Abstract

The invention discloses a potassium feldspar ceramic sintering whiteness detection system and method, and relates to the technical field of ceramic detection.The method comprises the steps that a space coordinate system is constructed in a ceramic sintering furnace, a plurality of high-temperature industrial cameras are used for periodically capturing ceramic images, image pixel coordinates are converted into three-dimensional coordinates through preprocessing, and then ceramic pixel points are screened out; with reference to historical sintering data, drawing a time sequence chart of the ceramic surface whiteness changing along with time, determining a standard whiteness value of each monitoring period, and meanwhile, fitting a correlation function of the in-furnace temperature and the sintering time; extracting RGB color data from the ceramic image data of the current monitoring period, converting the RGB color data into a whiteness value, and establishing a real-time monitoring whiteness time sequence diagram; by comparing the actual whiteness with the standard whiteness, the ceramic pixel points are classified, the abnormal pixels are identified, and whether the temperature needs to be adjusted or whether the model is abnormal or not is intelligently judged according to the number and the type of the abnormal pixels, so that the stability and the consistency of the ceramic sintering quality are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of ceramic detection, in particular to a system and method for detecting the sintered whiteness of potassium feldspar ceramics. Background Art

[0002] As an important natural mineral, potassium feldspar is widely used in the fields of construction, ceramics, glass and enamel due to its unique chemical composition and physical properties. After melting at high temperature, potassium feldspar can dissolve some of the decomposition products of quartz and kaolin, and is an indispensable flux raw material in ceramic manufacturing. However, the quality of potassium feldspar ceramics, especially its whiteness, directly affects the aesthetics and market value of the final product.

[0003] In the ceramic production process, whiteness is one of the important indicators to measure the quality of ceramic materials. Whiteness not only reflects the material's ability to reflect light, but is also closely related to the material's composition, structure, sintering process, etc. The whiteness detection of potassium feldspar ceramics traditionally relies on manual visual inspection or simple measurement using a colorimeter, but these methods have problems such as strong subjectivity, low efficiency, and insufficient precision. With the development of computer vision and image processing technology, whiteness detection technology based on machine vision has gradually become a research hotspot.

[0004] Traditional potassium feldspar ceramic sintering whiteness detection methods are often unable to monitor the whiteness changes during the ceramic sintering process in real time. If serious abnormalities have occurred during the ceramic sintering process, then continuing to wait for the end of the ceramic sintering will result in a waste of time and equipment resources. In addition, traditional potassium feldspar ceramic sintering whiteness detection methods often only monitor the whiteness of certain areas, and the whiteness distribution of potassium feldspar ceramics after sintering may not be uniform. Monitoring only certain areas cannot fully reflect the overall whiteness status. This incomplete detection result may lead to misjudgment of the sintering quality of potassium feldspar ceramics. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the technical problems in the background technology, the present invention proposes a potassium feldspar ceramic sintering whiteness detection system and method. Based on computer vision technology, by setting a monitoring cycle and detecting the whiteness value of each ceramic pixel at the end of each monitoring cycle, timely and accurate identification of whiteness anomalies in the ceramic sintering process is achieved, thereby solving the problems raised in the background technology.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A method for detecting the sintered whiteness of potassium feldspar ceramics, comprising:

[0010] A spatial rectangular coordinate system is established in the ceramic sintering furnace; a plurality of high-temperature industrial cameras are arranged at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, pixel coordinates in the images are converted into three-dimensional coordinates in the spatial rectangular coordinate system, thereby screening out ceramic pixel points in the images;

[0011] Obtain historical sintering data from the ceramic preparation library, draw a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures based on the acquired data, and determine the standard whiteness value of each ceramic pixel at the end of each monitoring period; fit the correlation function between the furnace temperature and the total sintering time based on the historical sintering duration and temperature data;

[0012] For the image data of the ceramic surface obtained in each monitoring cycle, the RGB color data of all ceramic pixels in the image are extracted, the color data is converted into whiteness values ​​using the Hunter white value algorithm, and a whiteness monitoring time series graph is established; the whiteness value of each ceramic pixel is compared with the standard whiteness value under the corresponding sintering time, and all ceramic pixels in each cycle are classified; based on the number and type of abnormal pixels, it is determined whether a temperature adjustment instruction or a model abnormality instruction needs to be issued.

[0013] Specifically, a spatial rectangular coordinate system is established with the center of the bottom surface of the ceramic sintering furnace as the origin, the east direction as the positive direction of the X axis, the north direction as the positive direction of the Y axis, and the upward direction as the positive direction of the Z axis;

[0014] The specific determination of the designated position is as follows: obtaining the height h1 of the ceramic model and the height h2 of the bottom of the ceramic in the ceramic sintering furnace, obtaining the height h2 of the ceramic sintering furnace A high-temperature industrial camera is placed at each of the four intersections of the side wall at a height with the X-axis plane and the Y-axis plane, and the lenses of the high-temperature industrial cameras face the ceramic model in the east, west, north and south directions respectively; a high-temperature industrial camera is placed at the center of the established spatial rectangular coordinate system, and faces the ceramic model directly above the Z axis.

[0015] Specifically, a monitoring cycle is set, and whenever the monitoring cycle ends, a shooting signal is sent to all high-temperature industrial cameras in the ceramic sintering furnace to obtain images in the corresponding direction;

[0016] The pixel coordinates of each pixel in the image are converted into three-dimensional coordinates in the camera coordinate system, and the direction of the camera coordinate system is adjusted to be consistent with the direction of the established spatial rectangular coordinate system;

[0017] Based on the rotation matrix R and translation vector s of each high-temperature industrial camera, the coordinates of each pixel in the image taken by each high-temperature industrial camera in the camera coordinate system are converted to the coordinates in the established spatial rectangular coordinate system. For the high-temperature industrial camera facing due west, it faces due east. At this time, the rotation matrix is ​​the unit matrix and the translation vector is For the high-temperature industrial camera in the east, it faces due west and rotates 180° counterclockwise around the Y axis relative to the east. At this time, the rotation matrix is ​​obtained by rotating 180° around the Y axis, and the translation vector is For a high-temperature industrial camera facing due north, it faces due south and rotates 270° counterclockwise around the Z axis relative to the due east direction. At this time, the rotation matrix is ​​obtained by rotating 270° around the Z axis, and the translation vector is For a high-temperature industrial camera facing due south, it faces due north and rotates 90° counterclockwise around the Z axis relative to the due east. At this time, the rotation matrix is ​​obtained by rotating 90° around the Z axis, and the translation vector is For the high-temperature industrial camera directly below, it is facing upward. At this time, the rotation matrix is ​​the unit matrix and the translation vector is (0,0,0) T .

[0018] Specifically, historical sintering data of corresponding potassium feldspar ceramics are obtained from the ceramic preparation library, including temperature data during sintering and whiteness monitoring values ​​of each ceramic pixel; each ceramic pixel in the historical sintering data is matched one by one with the ceramic pixel of the ceramic to be sintered currently based on the three-dimensional coordinates;

[0019] Several historical sintering data of potassium feldspar ceramics that have passed the ceramic whiteness test after sintering are screened out from the acquired historical data, and a time series diagram of the whiteness value Wh and sintering time t of each ceramic pixel point on the surface of potassium feldspar ceramics at different sintering temperatures is drawn, wherein the abscissa of the data series diagram is the total sintering time, and the ordinate is the standard whiteness value of the ceramic pixel point; for multiple potassium feldspar ceramic sintering data at the same temperature, the multiple whiteness values ​​of the ceramic pixel points at the same coordinate position and at the same monitoring time point and without whiteness abnormality are averaged, and the result is used as the standard whiteness value data of the corresponding ceramic pixel point position and the corresponding sintering time;

[0020] After the drawing is completed, a time series diagram of the standard whiteness values ​​of all ceramic pixels at several different temperatures is obtained.

[0021] Specifically, based on the sintering duration data of potassium feldspar ceramics at different temperatures, the temperature T in the ceramic sintering furnace and the total sintering time t are plotted. total The time series diagram is used to fit the correlation function t between the furnace temperature T and the total sintering time ttotal based on the data in the time series diagram.total (T), determine the corresponding relationship between the furnace temperature and the total sintering time from the function graph.

[0022] Specifically, at the end of each monitoring cycle, the image data captured by each high-temperature industrial camera is obtained, and the R, G, and B values ​​at each ceramic pixel in the image are obtained using GIMP image processing software;

[0023] Normalize the R, G, and B values ​​of each ceramic pixel; convert the normalized R, G, and B values ​​of each ceramic pixel into X, Y, and Z stimulus values ​​based on the sRGB conversion matrix; calculate the relative values ​​of the X, Y, and Z stimulus values ​​at each ceramic pixel relative to the X, Y, and Z stimulus values ​​under the D65 light source; convert the relative values ​​in the XYZ color space at each ceramic pixel into Lab values; calculate the whiteness value of each ceramic pixel based on the L, a, and b values ​​of each ceramic pixel in the Lab color space, and the expression is: The above conversion method is encapsulated into a whiteness conversion function. When an image of a ceramic surface is received, the whiteness conversion function is called to convert the whiteness value of each pixel on the ceramic surface.

[0024] Further, based on the whiteness value of each ceramic pixel at the end of each monitoring cycle, a whiteness monitoring time series diagram of each ceramic pixel is established, where the horizontal axis is the duration of the sintering operation and the vertical axis is the whiteness value of each ceramic pixel;

[0025] Obtain the temperature value set for the potassium feldspar ceramic sintering operation, and obtain the time series graph of the standard whiteness values ​​at all ceramic pixels at the corresponding temperature;

[0026] At the end of each monitoring cycle, the whiteness value at each ceramic pixel is calculated and the whiteness value Wh at each ceramic pixel is calculated. i The standard whiteness value of the pixel at the corresponding sintering time In comparison, if Then the pixel is recorded as an insufficient abnormal pixel;

[0027] like Then the pixel is recorded as a qualified pixel;

[0028] like The pixel is recorded as an excessive abnormal pixel.

[0029] Furthermore, at the end of each monitoring cycle, all the ceramic pixels on the ceramic surface are divided into three categories based on the size of the whiteness value, and the number of the three categories of ceramic pixels is counted respectively, and the number of insufficient abnormal pixels is recorded as n1, the number of qualified pixels is recorded as n2, and the number of excessive abnormal pixels is recorded as n3;

[0030] like No adjustment instructions will be triggered;

[0031] like The adjustment instruction is judged based on the quantitative relationship between the two types of abnormal pixels;

[0032] like The current monitoring cycle is marked as an abnormal cycle. If two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent.

[0033] Specifically, if The judgment of the adjustment instruction is made based on the quantitative relationship between the two types of abnormal pixels, specifically:

[0034] If n3<n1, based on the corresponding relationship between the furnace temperature and the total sintering time, if the higher the furnace temperature, the shorter the total sintering time, then the furnace temperature needs to be lowered at this time. Specifically, obtain the insufficient abnormal pixel points closest to the standard whiteness value in turn, calculate the difference between the whiteness value of these abnormal pixel points and the standard whiteness value in the current monitoring period, and add the absolute value TWh1 of the difference between the standard whiteness value of the abnormal pixel points corresponding to the current monitoring period and the standard whiteness value of the abnormal pixel points corresponding to the next monitoring period to obtain the whiteness value TWh2 to be adjusted; obtain the total sintering time Tt1 at the current temperature, based on the proportional formula Calculate the adjusted total sintering time Tt2, and calculate the temperature value T when the total sintering time is Tt2 based on the function ttotal(T) of the furnace temperature and the total sintering time. ad ; At this time, calculate in sequence when the temperature is adjusted to T ad After that, the whiteness value of each qualified pixel in the next monitoring cycle is calculated in descending order until the whiteness value is still within The number of qualified pixels that have been calculated is n5, and then the number of abnormal pixels in the next monitoring cycle is calculated. next , specific n next =n1-n4+n5-1+n3, where n4 represents the n4th insufficient abnormal pixel closest to the standard whiteness value; calculate the number of abnormal pixels in the next monitoring cycle after temperature adjustment of all insufficient abnormal pixels one by one next If there are insufficient abnormal pixels, after the temperature adjustment, the number of abnormal pixels in the next monitoring cycle If the temperature is adjusted, select n next The minimum temperature adjustment strategy is used, and the temperature in the ceramic sintering furnace is adjusted to this temperature; if there is no temperature adjustment for insufficient abnormal pixels, the number of abnormal pixels in the next monitoring cycle If the current monitoring cycle is marked as an abnormal cycle, if two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent;

[0035] If n3≥n1, the same analysis method as when n3<n1 is used to determine whether to adjust the temperature in the ceramic sintering furnace.

[0036] A potassium feldspar ceramic sintering whiteness detection system, comprising:

[0037] The image data acquisition and processing module establishes a spatial rectangular coordinate system in the ceramic sintering furnace; multiple high-temperature industrial cameras are arranged at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, the pixel coordinates in the images are converted into three-dimensional coordinates in the spatial rectangular coordinate system, thereby screening out the ceramic pixel points in the images;

[0038] The historical data acquisition and processing module acquires historical sintering data from the ceramic preparation library, draws a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures, and determines the standard whiteness value of each ceramic pixel at the end of each monitoring cycle; based on the historical sintering duration and temperature data, fits the correlation function between the furnace temperature and the total sintering time;

[0039] The whiteness conversion and monitoring module extracts the RGB color data of all ceramic pixels in the image from the image data of the ceramic surface acquired in each monitoring cycle, converts the color data into whiteness values ​​using the Hunter white value algorithm, and establishes a whiteness monitoring time series diagram;

[0040] The whiteness analysis and processing module compares the whiteness value of each ceramic pixel with the standard whiteness value under the corresponding sintering time, and classifies all ceramic pixels in each cycle; based on the number and type of abnormal pixels, it determines whether it is necessary to issue a temperature adjustment instruction or a model abnormality instruction.

[0041] (III) Beneficial effects

[0042] The present invention provides a potassium feldspar ceramic sintering whiteness detection system and method, which has the following beneficial effects:

[0043] 1. By establishing a spatial rectangular coordinate system and arranging multiple high-temperature industrial cameras in the ceramic sintering furnace, the sintering process of potassium feldspar ceramics can be monitored in all directions and at multiple angles. The depth information of each pixel point is obtained through the depth sensor, and the camera's internal reference matrix, optical center coordinates and other parameters are combined to achieve accurate conversion of image pixel coordinates to three-dimensional coordinates in the spatial rectangular coordinate system, which is convenient for subsequent matching with pixel points in historical data. Based on image recognition technology, the ceramic area is identified and ceramic pixel points are screened, providing strong data support for subsequent whiteness calculation and analysis.

[0044] 2. By obtaining the historical sintering data of potassium feldspar ceramics from the ceramic preparation library, and drawing a time series diagram of the whiteness change of each pixel point on the ceramic surface at different sintering temperatures based on these data, this method can comprehensively and accurately reflect the whiteness change of potassium feldspar ceramics during the sintering process; by determining the standard whiteness value of each ceramic pixel point at the end of each monitoring cycle, this method provides a reliable reference standard for subsequent whiteness detection; by fitting the correlation function between the furnace temperature and the total sintering time, the intrinsic connection between the furnace temperature and the total sintering time is further revealed, and a theoretical basis is provided for the subsequent adjustment of the furnace temperature;

[0045] 3. The Hunter white value algorithm is used to accurately convert color data into whiteness values, thereby establishing a detailed whiteness monitoring time series diagram and improving the accuracy of whiteness conversion; by carefully comparing the whiteness values ​​of all ceramic pixels with the standard whiteness values ​​under the corresponding sintering time, and accurately identifying and classifying all ceramic pixels in each cycle, this provides a solid data foundation for the subsequent temperature adjustment strategy, making the adjustment more accurate and efficient;

[0046] 4. According to the number and type of abnormal pixels, it is intelligently determined whether it is necessary to issue a temperature adjustment instruction or a model abnormality instruction; this intelligent decision-making mechanism not only avoids unnecessary adjustment operations, but also ensures the stability and continuity of the sintering process, thereby significantly improving the sintering quality and production efficiency of potassium feldspar ceramics; the above-mentioned whiteness detection method is carried out at the end of each monitoring cycle, which improves the timeliness of whiteness detection and avoids unnecessary waste of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A flow chart of the steps of a method for detecting the sintered whiteness of potassium feldspar ceramics provided by the present invention;

[0048] Figure 2 The present invention provides a schematic structural diagram of a potassium feldspar ceramic sintering whiteness detection system. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] refer to Figure 1 The present invention provides a method for detecting the sintered whiteness of potassium feldspar ceramics, comprising:

[0051] Step 1: Establish a spatial rectangular coordinate system in the ceramic sintering furnace; arrange multiple high-temperature industrial cameras at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, convert the pixel coordinates in the images into three-dimensional coordinates in the spatial rectangular coordinate system, and then screen out the ceramic pixel points in the images;

[0052] The step one includes the following steps:

[0053] Step 101, placing the potassium feldspar ceramic model into a ceramic sintering furnace, establishing a spatial rectangular coordinate system with the center of the bottom surface of the ceramic sintering furnace as the origin, the east direction as the positive direction of the X-axis, the north direction as the positive direction of the Y-axis, and the upward direction as the positive direction of the Z-axis; fixing the potassium feldspar ceramic model just above the bottom surface of the ceramic sintering furnace, that is, the center of the circle formed by its bottom edge is in the positive direction of the Z-axis;

[0054] Step 102: placing a plurality of high-temperature industrial cameras in the ceramic sintering furnace to obtain the surface image of the potassium feldspar ceramic model during the potassium feldspar ceramic sintering process; the method for determining the placement position of the high-temperature industrial cameras is specifically as follows: obtaining the height h1 of the ceramic model and the height h2 of the bottom of the ceramic in the ceramic sintering furnace, obtaining the height of the ceramic sintering furnace At four intersections of the side wall at the height and the X-axis plane and the Y-axis plane, a high-temperature industrial camera is placed at each of the four intersections, and the lenses of the high-temperature industrial cameras face the ceramic model in the east, west, north and south directions respectively; a high-temperature industrial camera is placed at the center of the spatial rectangular coordinate system established in step 101, and faces the ceramic model directly above the Z axis;

[0055] Step 103, setting a monitoring cycle, and sending a shooting signal to all high-temperature industrial cameras in the ceramic sintering furnace at the end of each monitoring cycle to obtain images in the corresponding direction; performing preprocessing operations on the obtained images, specifically including image denoising, image correction and image enhancement operations, wherein image denoising is used to remove noise caused by interference factors such as high temperature and smoke in the sintering furnace, image correction is used to correct deformation or distortion caused by factors such as the installation position and angle of the camera and lens distortion, and image enhancement is used to highlight key information in the image, which is the highlighting of each pixel in the image;

[0056] Step 104: The high temperature industrial camera is integrated with a depth sensor, that is, the image captured by the high temperature industrial camera contains the depth information of each pixel; obtain the internal parameter matrix K and the optical center coordinates (c u ,c v ), the focal length f on the x-axis and y-axis of the image u and f v , and the pixel coordinates (u, v) and depth value d of each pixel in the image captured by each high-temperature industrial camera, wherein the optical center coordinates are the coordinates at the center of the image captured by the camera; the pixel coordinates of each pixel in the image are converted into three-dimensional coordinates in the camera coordinate system, and the direction of the camera coordinate system is adjusted to be consistent with the direction of the spatial rectangular coordinate system established in step 101;

[0057] For high-temperature industrial cameras facing due east and due west, the conversion expression is:

[0058]

[0059] For high-temperature industrial cameras facing due north and due south, the conversion expression is:

[0060]

[0061] For a high-temperature industrial camera facing upward, the conversion expression is:

[0062]

[0063] Based on the rotation matrix R and the translation vector s of the position of each high-temperature industrial camera, the coordinates of each pixel in the image taken by each high-temperature industrial camera in the camera coordinate system are converted to the coordinates in the spatial rectangular coordinate system established in step 101. The conversion expression is:

[0064]

[0065] Among them, for the high-temperature industrial camera in the west, it faces the east direction. At this time, the rotation matrix is ​​the unit matrix, and the translation vector is in() T The symbol indicates the transposition operation of the vector. For the high-temperature industrial camera in the east, it faces the west and rotates 180° counterclockwise around the Y axis relative to the east. At this time, the rotation matrix is ​​obtained by rotating 180° around the Y axis, and the translation vector is For a high-temperature industrial camera facing due north, it faces due south and rotates 270° counterclockwise around the Z axis relative to the due east direction. At this time, the rotation matrix is ​​obtained by rotating 270° around the Z axis, and the translation vector is For a high-temperature industrial camera facing due south, it faces due north and rotates 90° counterclockwise around the Z axis relative to the due east. At this time, the rotation matrix is ​​obtained by rotating 90° around the Z axis, and the translation vector is For the high-temperature industrial camera directly below, it is facing upward. At this time, the rotation matrix is ​​the unit matrix and the translation vector is (0,0,0) T ;

[0066] Step 105, based on the five images acquired at the end of each cycle and the three-dimensional coordinates of each pixel in the image; based on image recognition technology, the ceramic area in each image is identified, and all the pixels in the ceramic area are recorded as ceramic pixels.

[0067] When using, combine the contents in steps 101 to 105:

[0068] By establishing a spatial rectangular coordinate system and arranging multiple high-temperature industrial cameras in the ceramic sintering furnace, the sintering process of potassium feldspar ceramics can be monitored in all directions and at multiple angles. The depth information of each pixel point is obtained through a depth sensor, and the camera's internal reference matrix, optical center coordinates and other parameters are combined to achieve accurate conversion of image pixel coordinates to three-dimensional coordinates in the spatial rectangular coordinate system, which is convenient for subsequent matching with pixel points in historical data. Based on image recognition technology, the ceramic area is identified and ceramic pixel points are screened, providing strong data support for subsequent whiteness calculation and analysis.

[0069] Step 2: Obtain historical sintering data from the ceramic preparation library, draw a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures based on the acquired data, and determine the standard whiteness value of each ceramic pixel at the end of each monitoring cycle; based on the historical sintering duration and temperature data, fit the correlation function between the furnace temperature and the total sintering time;

[0070] The step 2 includes the following steps:

[0071] Step 201, obtaining historical sintering data of corresponding potassium feldspar ceramics from a ceramic preparation library, including temperature data during sintering and whiteness monitoring values ​​of each ceramic pixel point; in the historical sintering data, the ceramic pixel points are also established and obtained according to the method in step 1, and the placement position and orientation of each surface of the ceramic are the same during each sintering, so as to obtain a plurality of ceramic pixel points at the same coordinate position, and then, based on the three-dimensional coordinates, each ceramic pixel point in the historical sintering data is matched one by one with the ceramic pixel point of the ceramic to be sintered currently;

[0072] From the data obtained above, several historical sintering data of potassium feldspar ceramics that have passed the ceramic whiteness test after sintering are screened out. Based on the historical sintering data of multiple potassium feldspar ceramics (including the changes in the whiteness monitoring value of each ceramic pixel point), a time series diagram of the whiteness value Wh and the sintering time t of each ceramic pixel point on the surface of the potassium feldspar ceramic at different sintering temperatures is plotted, wherein the abscissa of the data series diagram is the total sintering time, and the ordinate is the standard whiteness value of the ceramic pixel point; the standard whiteness value represents the whiteness value of each ceramic pixel point monitored at a certain monitoring time point in the historical sintering data of potassium feldspar ceramics when no abnormality is found, such as the judgment in the subsequent step 303; for multiple potassium feldspar ceramics at the same temperature, the whiteness value Wh of each ceramic pixel point on the surface of the potassium feldspar ceramic is plotted against the sintering time t. For the sintering data of potassium feldspar ceramics, the whiteness values ​​of the ceramic pixels at the same coordinate position and at the same monitoring time point without whiteness abnormality are averaged to obtain the standard whiteness value data corresponding to the ceramic pixel position and the corresponding monitoring time, i.e., the sintering time in the above-mentioned time series diagram; for example: if there are 3 groups of potassium feldspar ceramic historical sintering data at the same specified temperature in the potassium feldspar ceramic historical sintering data, in these data, at the end of the 5th monitoring cycle, a specified pixel point on the ceramic surface has an abnormal whiteness value once and no abnormality twice, then the average of the two whiteness values ​​without abnormality is taken as the standard whiteness value at the end of the 5th monitoring cycle at the specified temperature;

[0073] Based on the above operation, after the drawing is completed, a time series diagram of the standard whiteness values ​​of all ceramic pixels at different temperatures is obtained;

[0074] Step 202: based on the sintering duration data of potassium feldspar ceramics at different temperatures, plot the temperature T in the ceramic sintering furnace and the total sintering time t total The time series diagram is used to fit the correlation function t between the furnace temperature T and the total sintering time ttotal based on the data in the time series diagram. total (T), from the function graph, we can get the corresponding relationship between the furnace temperature and the total sintering time, that is, the higher the furnace temperature, the longer or shorter the total sintering time.

[0075] When using, combine the contents in steps 201 to 201:

[0076] By obtaining the historical sintering data of potassium feldspar ceramics from the ceramic preparation library, and based on these data, a time series diagram of the whiteness change of each pixel point on the ceramic surface at different sintering temperatures is drawn. This method can comprehensively and accurately reflect the whiteness change of potassium feldspar ceramics during the sintering process; by determining the standard whiteness value of each ceramic pixel point at the end of each monitoring cycle, this method provides a reliable reference standard for subsequent whiteness detection; by fitting the correlation function between the furnace temperature and the total sintering time, the intrinsic relationship between the furnace temperature and the total sintering time is further revealed, and a theoretical basis is provided for the subsequent adjustment of the furnace temperature;

[0077] Step 3: For the image data of the ceramic surface obtained in each monitoring cycle, extract the RGB color data of all ceramic pixels in the image, convert the color data into whiteness value using Hunter white value algorithm, and establish a whiteness monitoring time series diagram; compare the whiteness value of each ceramic pixel with the standard whiteness value under the corresponding sintering time, and classify all ceramic pixels in each cycle; determine whether it is necessary to issue a temperature adjustment instruction or a model abnormality instruction based on the number and type of abnormal pixels;

[0078] The step three includes the following steps:

[0079] Step 301, after starting the sintering operation of potassium feldspar ceramics, based on the set monitoring cycle length, at the end of each monitoring cycle, the image data captured by each high-temperature industrial camera is obtained, the ceramic pixel points in step 105 are screened for the image data, and the R, G, and B values ​​at each ceramic pixel point are obtained using GIMP image processing software;

[0080] Step 302: Based on the Hunter white value algorithm, the R, G, and B values ​​of each ceramic pixel are converted into a specific whiteness value. The specific conversion method is:

[0081] Step 3021: normalize the R, G, and B values ​​of each ceramic pixel. The expression is:

[0082]

[0083] Step 3022: based on the sRGB conversion matrix, the normalized R, G, B values ​​of each ceramic pixel are converted into X, Y, Z tristimulus values, expressed as:

[0084]

[0085] Where H represents the sRGB conversion matrix, and

[0086] Step 3023, calculate the relative values ​​of the X, Y, and Z tristimulus values ​​at each ceramic pixel relative to the X, Y, and Z tristimulus values ​​under the D65 light source, and the expression is:

[0087]

[0088] Among them, X n , Y n , Z n They represent the X, Y, and Z tristimulus values ​​under the D65 light source. The D65 light source is a standard light source that represents natural daylight (average daylight). It can better simulate the color display under the daylight environment. Therefore, in many fields, the D65 light source is widely used as a standard light source for color evaluation and comparison;

[0089] Step 3024: convert the relative value in the XYZ color space at each ceramic pixel into a Lab value, wherein the Lab value is derived based on the Lab color space to accurately describe and quantify the color; the Lab value includes an L value, an a value, and a b value, wherein the L value represents the brightness of the color, i.e., the lightness, the a value represents the position of the color on the red-green axis, and the b value represents the position of the color on the yellow-blue axis; the specific conversion formula is:

[0090]

[0091] Wherein, function f(x) represents a nonlinear conversion function that converts relative values ​​in XYZ color space (relative to D65 light source) into L, a, b values ​​in Lab color space, and

[0092] Step 3025: Calculate the whiteness value (Hunter whiteness value) of each ceramic pixel based on the L, a, and b values ​​of each ceramic pixel in the Lab color space. The expression is: In the historical sintering data stored in the ceramic preparation library, the whiteness value of each pixel is also calculated according to step 302, and the conversion method in step 302 is encapsulated into a whiteness conversion function. When the image of the ceramic surface is received, the whiteness conversion function is called to convert the whiteness value of each pixel on the ceramic surface;

[0093] Step 303: at the end of each monitoring cycle, the whiteness value of each ceramic pixel is calculated based on the image captured by the high-temperature industrial camera; a whiteness monitoring time series diagram of each ceramic pixel is established, where the horizontal axis is the duration of the sintering operation and the vertical axis is the whiteness value of each ceramic pixel;

[0094] Obtain the temperature value set for the potassium feldspar ceramic sintering operation, and obtain the time series graph of the standard whiteness values ​​at all ceramic pixels at the corresponding temperature;

[0095] Starting from the sintering operation of potassium feldspar ceramics, the whiteness value at each ceramic pixel is calculated at the end of each monitoring cycle, and the whiteness value Wh at each ceramic pixel is calculated. i The standard whiteness value of the pixel at the corresponding sintering time In comparison, if It means that at the end of the current monitoring cycle, the whiteness value Wh at the i-th ceramic pixel i Less than the standard whiteness value under the corresponding sintering time That is, the whiteness at this point on the surface of the potassium feldspar ceramic is insufficient, and the pixel point is recorded as an insufficient abnormal pixel point;

[0096] like It means that at the end of the current monitoring cycle, the whiteness value Wh at the i-th ceramic pixel i Standard whiteness value at the corresponding sintering time If the pixel is within the error range of , that is, the whiteness at this point on the surface of the potassium feldspar ceramic is qualified, the pixel is recorded as a qualified pixel, and also as a ceramic pixel without abnormality;

[0097] like It means that at the end of the current monitoring cycle, the whiteness value Wh at the i-th ceramic pixel i Greater than the standard whiteness value under the corresponding sintering time That is, the whiteness at this point on the surface of the potassium feldspar ceramic is excessive, and the pixel point is recorded as an excessive abnormal pixel point;

[0098] Step 304: At the end of each monitoring cycle, all the ceramic pixels on the ceramic surface are divided into three categories according to step 303; the number of the three categories of ceramic pixels is counted respectively, and the number of insufficient abnormal pixels is recorded as n1, the number of qualified pixels is recorded as n2, and the number of excessive abnormal pixels is recorded as n3;

[0099] Determine the relationship between n1, n2 and n3, if This means that at the end of the current monitoring cycle, the number of abnormal pixels is within the normal range and no adjustment instructions are triggered;

[0100] like This means that at the end of the current monitoring cycle, the number of abnormal pixels is not within the normal range, but the number of abnormal pixels can be adjusted to the normal range by adjusting the temperature. At this time, the adjustment instruction is judged based on the quantitative relationship between the two types of abnormal pixels, specifically:

[0101] If n3<n1, it means that at the end of the current monitoring cycle, the number of insufficient abnormal pixels is greater than the number of excessive abnormal pixels. At this time, it is necessary to determine whether the temperature in the ceramic sintering furnace can be adjusted; based on the corresponding relationship between the furnace temperature and the total sintering time in step 203, if the higher the furnace temperature, the shorter the total sintering time, then the furnace temperature needs to be lowered to slow down the rate of decrease in the whiteness of the ceramic surface, and then adjust these insufficient abnormal pixels, specifically: obtain the insufficient abnormal pixels closest to the standard whiteness value in turn, calculate the difference between the whiteness value of the abnormal pixel in the current monitoring cycle and the standard whiteness value, and add it to the absolute value TWh1 of the difference between the standard whiteness value of the abnormal pixel in the current monitoring cycle and the standard whiteness value of the abnormal pixel in the next monitoring cycle, to obtain the whiteness value TWh2 to be adjusted; obtain the total sintering time Tt1 at the current temperature, based on the proportional formula Calculate the adjusted total sintering time Tt2, and calculate the temperature value T when the total sintering time is Tt2 based on the function ttotal(T) of the furnace temperature and the total sintering time. ad ; At this time, the qualified pixels are calculated in sequence and the temperature is adjusted to T ad After that, the whiteness value of each qualified pixel in the next monitoring cycle is calculated in descending order until the whiteness value is still within The number of qualified pixels that have been calculated is n5, and then the number of abnormal pixels in the next monitoring cycle is calculated. next , specific n next =n1-n4+n5-1+n3, where n4 represents the n4th insufficient abnormal pixel closest to the standard whiteness value; calculate the number of abnormal pixels in the next monitoring cycle after temperature adjustment of all insufficient abnormal pixels one by one next If there are insufficient abnormal pixels, after the temperature adjustment, the number of abnormal pixels in the next monitoring cycle If the temperature is adjusted, select n next The minimum temperature adjustment strategy is used, and the temperature in the ceramic sintering furnace is adjusted to this temperature; if there is no temperature adjustment for insufficient abnormal pixels, the number of abnormal pixels in the next monitoring cycle If the current monitoring cycle is marked as an abnormal cycle, if two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent and the potassium feldspar model needs to be rebuilt;

[0102] If n3≥n1, it means that at the end of the current monitoring cycle, the number of excess abnormal pixels is not less than the number of excess abnormal pixels. At this time, the above analysis method, that is, the same analysis method as when n3<n1, is used to determine whether the temperature in the ceramic sintering furnace can be adjusted. The specific method is: based on the corresponding relationship between the temperature in the furnace and the total sintering time, if the temperature in the furnace is higher, the total sintering time is shorter, then the temperature in the furnace needs to be increased at this time, specifically: obtain the excess abnormal pixels closest to the standard whiteness value in turn, calculate the difference between the whiteness value of these abnormal pixels in the current monitoring cycle and the standard whiteness value, and add the absolute value TWh1 of the difference between the standard whiteness value of the abnormal pixel corresponding to the current monitoring cycle and the standard whiteness value of the abnormal pixel corresponding to the next monitoring cycle to obtain the whiteness value TWh2 to be adjusted; obtain the total sintering time Tt1 at the current temperature, based on the proportional formula Calculate the adjusted total sintering time Tt2, and calculate the temperature value T when the total sintering time is Tt2 based on the function ttotal(T) of the furnace temperature and the total sintering time. ad ; At this time, calculate in sequence when the temperature is adjusted to T ad After that, the whiteness value of each qualified pixel in the next monitoring cycle is calculated in ascending order until the whiteness value is still within The number of qualified pixels that have been calculated is n5, and then the number of abnormal pixels in the next monitoring cycle is calculated. next , specific n next =n1+n5-1+n3-n4, where n4 represents the n4th excessive abnormal pixel closest to the standard whiteness value; calculate the number of abnormal pixels in the next monitoring cycle after temperature adjustment of all excessive abnormal pixels one by one n next If there are excessive abnormal pixels, the number of abnormal pixels in the next monitoring cycle will be If the temperature is adjusted, select n next The minimum temperature adjustment strategy is used, and the temperature in the ceramic sintering furnace is adjusted to this temperature; if there is no temperature adjustment for excessive abnormal pixels, the number of abnormal pixels in the next monitoring cycle If the current monitoring cycle is marked as an abnormal cycle, if two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent;

[0103] like It means that at the end of the current monitoring cycle, the number of abnormal pixels is not within the normal range and the number of abnormal pixels cannot be adjusted to the normal range by adjusting the temperature. The current monitoring cycle is marked as an abnormal cycle. If two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent and the potassium feldspar model needs to be rebuilt.

[0104] When used, combine the contents in steps 301 to 304:

[0105] The Hunter white value algorithm is used to accurately convert color data into whiteness values, thereby establishing a detailed whiteness monitoring time series diagram and improving the accuracy of whiteness conversion; by carefully comparing the whiteness values ​​of all ceramic pixels with the standard whiteness values ​​under the corresponding sintering time, and accurately identifying and classifying all ceramic pixels in each cycle, this provides a solid data foundation for the subsequent temperature adjustment strategy, making the adjustment more accurate and efficient;

[0106] According to the number and type of abnormal pixels, it is intelligently determined whether a temperature adjustment instruction or a model abnormality instruction needs to be issued; this intelligent decision-making mechanism not only avoids unnecessary adjustment operations, but also ensures the stability and continuity of the sintering process, thereby significantly improving the sintering quality and production efficiency of potassium feldspar ceramics; the above-mentioned whiteness detection method is carried out at the end of each monitoring cycle, which improves the timeliness of whiteness detection and avoids unnecessary waste of resources.

[0107] refer to Figure 2 The present invention also provides a potassium feldspar ceramic sintering whiteness detection system, comprising:

[0108] The image data acquisition and processing module establishes a spatial rectangular coordinate system in the ceramic sintering furnace; multiple high-temperature industrial cameras are arranged at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, the pixel coordinates in the images are converted into three-dimensional coordinates in the spatial rectangular coordinate system, thereby screening out the ceramic pixel points in the images;

[0109] The historical data acquisition and processing module acquires historical sintering data from the ceramic preparation library, draws a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures, and determines the standard whiteness value of each ceramic pixel at the end of each monitoring cycle; based on the historical sintering duration and temperature data, fits the correlation function between the furnace temperature and the total sintering time;

[0110] The whiteness conversion and monitoring module extracts the RGB color data of all ceramic pixels in the image from the image data of the ceramic surface acquired in each monitoring cycle, converts the color data into whiteness values ​​using the Hunter white value algorithm, and establishes a whiteness monitoring time series diagram;

[0111] The whiteness analysis and processing module compares the whiteness value of each ceramic pixel with the standard whiteness value under the corresponding sintering time, and classifies all ceramic pixels in each cycle; based on the number and type of abnormal pixels, it determines whether it is necessary to issue a temperature adjustment instruction or a model abnormality instruction.

[0112] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer storage medium or transmitted via a computer storage medium.

[0113] Computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer storage media can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state drives (SSDs)).

[0114] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for detecting the sintered whiteness of potassium feldspar ceramics, characterized in that: include: A spatial rectangular coordinate system is established in the ceramic sintering furnace; a plurality of high-temperature industrial cameras are arranged at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, pixel coordinates in the images are converted into three-dimensional coordinates in the spatial rectangular coordinate system, thereby screening out ceramic pixel points in the images; Obtain historical sintering data from the ceramic preparation library, draw a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures based on the acquired data, and determine the standard whiteness value of each ceramic pixel at the end of each monitoring period; fit the correlation function between the furnace temperature and the total sintering time based on the historical sintering duration and temperature data; For the image data of the ceramic surface obtained in each monitoring cycle, the RGB color data of all ceramic pixels in the image are extracted, the color data is converted into whiteness values ​​using the Hunter white value algorithm, and a whiteness monitoring time series graph is established; the whiteness value of each ceramic pixel is compared with the standard whiteness value under the corresponding sintering time, and all ceramic pixels in each cycle are classified; based on the number and type of abnormal pixels, it is determined whether a temperature adjustment instruction or a model abnormality instruction needs to be issued.

2. A method for detecting sintered whiteness of potassium feldspar ceramics according to claim 1, characterized in that: A spatial rectangular coordinate system is established with the center of the bottom surface of the ceramic sintering furnace as the origin, the east direction as the positive direction of the X axis, the north direction as the positive direction of the Y axis, and the upward direction as the positive direction of the Z axis; The specific determination of the designated position is as follows: obtaining the height h1 of the ceramic model and the height h2 of the bottom of the ceramic in the ceramic sintering furnace, obtaining the height h2 of the ceramic sintering furnace A high-temperature industrial camera is placed at each of the four intersections of the side wall at a height of h1+h2 and the X-axis plane and the Y-axis plane, and the lenses of the high-temperature industrial cameras face the ceramic model in the east, west, north and south directions respectively; a high-temperature industrial camera is placed at the center of the established spatial rectangular coordinate system, and faces the ceramic model directly above the Z axis.

3. A method for detecting the sintered whiteness of potassium feldspar ceramics as claimed in claim 2, characterized in that: Set a monitoring cycle, and when the monitoring cycle ends, send a shooting signal to all high-temperature industrial cameras in the ceramic sintering furnace to obtain images in the corresponding direction; The pixel coordinates of each pixel in the image are converted into three-dimensional coordinates in the camera coordinate system, and the direction of the camera coordinate system is adjusted to be consistent with the direction of the established spatial rectangular coordinate system; Based on the rotation matrix R and translation vector s of each high-temperature industrial camera, the coordinates of each pixel in the image taken by each high-temperature industrial camera in the camera coordinate system are converted to the coordinates in the established spatial rectangular coordinate system. For the high-temperature industrial camera facing due west, it faces due east. At this time, the rotation matrix is ​​the unit matrix and the translation vector is For the high-temperature industrial camera in the east, it faces due west and rotates 180° counterclockwise around the Y axis relative to the east. At this time, the rotation matrix is ​​obtained by rotating 180° around the Y axis, and the translation vector is For a high-temperature industrial camera facing due north, it faces due south and rotates 270° counterclockwise around the Z axis relative to the due east direction. At this time, the rotation matrix is ​​obtained by rotating 270° around the Z axis, and the translation vector is For a high-temperature industrial camera facing due south, it faces due north and rotates 90° counterclockwise around the Z axis relative to the due east. At this time, the rotation matrix is ​​obtained by rotating 90° around the Z axis, and the translation vector is For the high-temperature industrial camera directly below, it is facing upward. At this time, the rotation matrix is ​​the unit matrix and the translation vector is (0,0,0) T .

4. A method for detecting the sintered whiteness of potassium feldspar ceramics as claimed in claim 3, characterized in that: Obtain the historical sintering data of the corresponding potassium feldspar ceramic from the ceramic preparation library, including the temperature data during sintering and the whiteness monitoring value of each ceramic pixel; and correspond each ceramic pixel in the historical sintering data to the ceramic pixel of the ceramic to be sintered based on the three-dimensional coordinates; Several historical sintering data of potassium feldspar ceramics that have passed the ceramic whiteness test after sintering are screened out from the acquired historical data, and a time series diagram of the whiteness value Wh and sintering time t of each ceramic pixel point on the surface of potassium feldspar ceramics at different sintering temperatures is drawn, wherein the abscissa of the data series diagram is the total sintering time, and the ordinate is the standard whiteness value of the ceramic pixel point; for multiple potassium feldspar ceramic sintering data at the same temperature, the multiple whiteness values ​​of the ceramic pixel points at the same coordinate position and at the same monitoring time point and without whiteness abnormality are averaged, and the result is used as the standard whiteness value data of the corresponding ceramic pixel point position and the corresponding sintering time; After the drawing is completed, a time series diagram of the standard whiteness values ​​of all ceramic pixels at several different temperatures is obtained.

5. A method for detecting sintered whiteness of potassium feldspar ceramics as claimed in claim 4, characterized in that: Based on the sintering duration data of potassium feldspar ceramics at different temperatures, the temperature T in the ceramic sintering furnace and the total sintering time t are plotted. total The time series diagram is used to fit the correlation function t between the furnace temperature T and the total sintering time ttotal based on the data in the time series diagram. total (T), determine the corresponding relationship between the furnace temperature and the total sintering time from the function graph.

6. A method for detecting sintered whiteness of potassium feldspar ceramics as claimed in claim 5, characterized in that: At the end of each monitoring cycle, the image data captured by each high-temperature industrial camera is obtained, and the R, G, and B values ​​of each ceramic pixel in the image are obtained using GIMP image processing software; Normalize the R, G, and B values ​​of each ceramic pixel; convert the normalized R, G, and B values ​​of each ceramic pixel into X, Y, and Z stimulus values ​​based on the sRGB conversion matrix; calculate the relative values ​​of the X, Y, and Z stimulus values ​​at each ceramic pixel relative to the X, Y, and Z stimulus values ​​under the D65 light source; convert the relative values ​​in the XYZ color space at each ceramic pixel into Lab values; calculate the whiteness value of each ceramic pixel based on the L, a, and b values ​​of each ceramic pixel in the Lab color space, and the expression is: The above conversion method is encapsulated into a whiteness conversion function. When an image of a ceramic surface is received, the whiteness conversion function is called to convert the whiteness value of each pixel on the ceramic surface.

7. A method for detecting sintered whiteness of potassium feldspar ceramics as claimed in claim 6, characterized in that: Based on the whiteness value of each ceramic pixel at the end of each monitoring cycle, a whiteness monitoring time series diagram of each ceramic pixel is established, where the horizontal axis is the duration of the sintering operation and the vertical axis is the whiteness value of each ceramic pixel; Obtain the temperature value set for the potassium feldspar ceramic sintering operation, and obtain the time series graph of the standard whiteness values ​​at all ceramic pixels at the corresponding temperature; At the end of each monitoring cycle, the whiteness value at each ceramic pixel is calculated and the whiteness value Wh at each ceramic pixel is calculated. i The standard whiteness value of the pixel at the corresponding sintering time In comparison, if Then the pixel is recorded as an insufficient abnormal pixel; like Then the pixel point is recorded as a qualified pixel point; like The pixel is recorded as an excessive abnormal pixel.

8. A method for detecting sintered whiteness of potassium feldspar ceramics as claimed in claim 7, characterized in that: At the end of each monitoring cycle, all the ceramic pixels on the ceramic surface are divided into three categories based on the size of the whiteness value, and the number of ceramic pixels in the three categories is counted respectively, and the number of insufficient abnormal pixels is recorded as n1, the number of qualified pixels is recorded as n2, and the number of excessive abnormal pixels is recorded as n3; like No adjustment instructions will be triggered; like The adjustment instruction is judged based on the quantitative relationship between the two types of abnormal pixels; like The current monitoring cycle is marked as an abnormal cycle. If two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent.

9. A method for detecting sintered whiteness of potassium feldspar ceramics as claimed in claim 8, characterized in that: like The judgment of the adjustment instruction is made based on the quantitative relationship between the two types of abnormal pixels, specifically: If n3<n1, based on the corresponding relationship between the furnace temperature and the total sintering time, if the higher the furnace temperature, the shorter the total sintering time, then the furnace temperature needs to be lowered at this time. Specifically, obtain the insufficient abnormal pixel points closest to the standard whiteness value in turn, calculate the difference between the whiteness value of these abnormal pixel points and the standard whiteness value in the current monitoring period, and add the absolute value TWh1 of the difference between the standard whiteness value of the abnormal pixel points corresponding to the current monitoring period and the standard whiteness value of the abnormal pixel points corresponding to the next monitoring period to obtain the whiteness value TWh2 to be adjusted; obtain the total sintering time Tt1 at the current temperature, based on the proportional formula Calculate the adjusted total sintering time Tt2, and calculate the temperature value T when the total sintering time is Tt2 based on the function ttotal(T) of the furnace temperature and the total sintering time. ad ; At this time, calculate in sequence when the temperature is adjusted to T ad After that, the whiteness value of each qualified pixel in the next monitoring cycle is calculated in descending order until the whiteness value is still within The number of qualified pixels that have been calculated is n5, and then the number of abnormal pixels in the next monitoring cycle is calculated. next , specifically: n next =n1-n4+n5-1+n3, where n4 represents the n4th insufficient abnormal pixel closest to the standard whiteness value; calculate the number of abnormal pixels in the next monitoring cycle after temperature adjustment of all insufficient abnormal pixels one by one next If there are insufficient abnormal pixels, after the temperature adjustment, the number of abnormal pixels in the next monitoring cycle If the temperature is adjusted, select n next The minimum temperature adjustment strategy is used, and the temperature in the ceramic sintering furnace is adjusted to this temperature; if there is no temperature adjustment for insufficient abnormal pixels, the number of abnormal pixels in the next monitoring cycle If the current monitoring cycle is marked as an abnormal cycle, if two consecutive monitoring cycles are marked as abnormal cycles, a model abnormality instruction is sent; If n3≥n1, the same analysis method as when n3<n1 is used to determine whether to adjust the temperature in the ceramic sintering furnace.

10. A potassium feldspar ceramic sintering whiteness detection system, characterized in that: include: The image data acquisition and processing module establishes a spatial rectangular coordinate system in the ceramic sintering furnace; multiple high-temperature industrial cameras are arranged at designated positions in the ceramic sintering furnace to periodically acquire potassium feldspar ceramic images in corresponding directions; after preprocessing the acquired images, the pixel coordinates in the images are converted into three-dimensional coordinates in the spatial rectangular coordinate system, thereby screening out the ceramic pixel points in the images; The historical data acquisition and processing module acquires historical sintering data from the ceramic preparation library, draws a time series diagram of the whiteness of each pixel on the ceramic surface at different sintering temperatures, and determines the standard whiteness value of each ceramic pixel at the end of each monitoring cycle; based on the historical sintering duration and temperature data, fits the correlation function between the furnace temperature and the total sintering time; The whiteness conversion and monitoring module extracts the RGB color data of all ceramic pixels in the image from the image data of the ceramic surface acquired in each monitoring cycle, converts the color data into whiteness values ​​using the Hunter white value algorithm, and establishes a whiteness monitoring time series diagram; The whiteness analysis and processing module compares the whiteness value of each ceramic pixel with the standard whiteness value under the corresponding sintering time, and classifies all ceramic pixels in each cycle; based on the number and type of abnormal pixels, it determines whether it is necessary to issue a temperature adjustment instruction or a model abnormality instruction.

Images