A ceramic color detection method and device

Abstract

This invention relates to the field of color detection technology, and discloses a method and apparatus for ceramic color detection, comprising: extracting multi-directional ceramic images of a preset ceramic sample; dividing the multi-directional ceramic images into grids; calculating the image sharpness of the multi-directional ceramic grid images; performing image enhancement processing on the multi-directional ceramic grid images based on the image sharpness; constructing a wavelength matrix based on wavelength attributes; generating a reflection matrix of the multi-directional ceramic grid images based on the wavelength matrix; mapping the reflection matrix and the wavelength matrix; generating a color spectral curve of the multi-directional ceramic grid images through a color spectral matrix; extracting color spectral features from the color spectral curve; calculating the correlation degree of the multi-directional ceramic grid images; and performing grid color aggregation on the multi-directional ceramic grid images based on the correlation degree and color spectral features to generate the color distribution of the ceramic sample. This invention can improve the accuracy of ceramic color detection.

Description

Technical Field

[0001] This invention relates to the field of color detection technology, and in particular to a method and apparatus for detecting the color of ceramics. Background Technology

[0002] With the development of industrialized production, modern ceramic manufacturing increasingly relies on chemical dyes and additives to improve the color stability and consistency of products. However, since the color characteristics of ceramic samples are affected by a variety of factors, it is necessary to conduct color testing on ceramic samples to ensure that they meet the required color standards.

[0003] Existing ceramic color detection technologies typically employ visual comparison or color chart comparison to perform color detection on ceramic samples. In practical applications, relying solely on visual inspection or color chart comparison is subjective and uncertain, and cannot provide objective data support. This may lead to an overly simplistic approach to ceramic color detection, resulting in low accuracy. Summary of the Invention

[0004] This invention provides a method and apparatus for detecting ceramic color, the main purpose of which is to solve the problem of low accuracy in detecting ceramic color.

[0005] To achieve the above objectives, the present invention provides a ceramic color detection method, comprising:

[0006] S1. Extract multi-directional ceramic images of a preset ceramic sample, divide the multi-directional ceramic images into grids to obtain multi-directional ceramic grid images, and calculate the image clarity of the multi-directional ceramic grid images using a preset image clarity algorithm.

[0007] S2. Perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes;

[0008] S3. Generate the reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix;

[0009] S4. Generate the color spectral curve of the multi-directional ceramic grid image through the color spectral matrix, and extract the color spectral features in the color spectral curve;

[0010] S5. Calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers; perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample; wherein calculating the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers includes:

[0011] S51. Extract the reflectance peak value from the color spectral features;

[0012] S52. Determine the grid region identifier of each grid in the multi-directional ceramic grid image according to the preset grid identifier;

[0013] S53. Calculate the correlation degree of the multi-directional ceramic grid image based on the peak reflectance and the grid region identifier, wherein the correlation degree calculation formula is:

[0014]

[0015] Where G is the correlation degree, u t To identify the peak reflectance value corresponding to the t-th grid region, ρ t1 To identify the wavelength color corresponding to the reflectivity peak of the t-th grid region, u p The p-th grid region is identified by its corresponding reflectance peak value, ρ. p2 Identify the wavelength color corresponding to the reflectivity peak of the p-th grid region.

[0016] Optionally, the extraction of multi-directional ceramic images of a preset ceramic sample includes:

[0017] Generate multi-angle view attributes of ceramic samples based on preset view directions and preset view angles;

[0018] Generate multi-angle ceramic view states of ceramic samples one by one according to the multi-angle view attributes;

[0019] A multi-directional ceramic image of the ceramic sample is generated based on the multi-directional ceramic view state.

[0020] Optionally, calculating the image sharpness of the multi-directional ceramic grid image using a preset image sharpness algorithm includes:

[0021] Extract the pixel attributes of the multi-directional ceramic mesh image;

[0022] The difference between each pixel in the multi-directional ceramic mesh image is calculated using a preset difference algorithm.

[0023] The image sharpness of the multi-directional ceramic mesh image is calculated using the following preset image sharpness algorithm based on the pixel attributes and the difference:

[0024]

[0025] Where Q is the image sharpness, M is the image row dimension in the pixel attribute, N is the image column dimension in the pixel attribute, and ΔF x Let ΔF be the difference between pixels (m,n) in the x-direction. y Let δ be the difference between pixels (m,n) in the y direction, and let δ be the sharpness optimization factor.

[0026] Optionally, the step of performing image enhancement processing on the multi-directional ceramic mesh image based on the image sharpness to obtain the target multi-directional ceramic mesh image includes:

[0027] When the image sharpness is less than or equal to a preset sharpness threshold, a sharpness parameter enhancement combination is generated based on preset contrast parameters, preset sharpening parameters, and preset resolution parameters.

[0028] The multi-directional ceramic mesh image is enhanced according to the aforementioned sharpness parameter enhancement combination to obtain a multi-directional ceramic mesh enhanced image;

[0029] Calculate the target sharpness of the multi-directional ceramic mesh enhanced image. When the target sharpness is less than or equal to a preset sharpness threshold, adjust the sharpness parameter enhancement combination and return to the step of performing image enhancement processing on the multi-directional ceramic mesh image according to the sharpness parameter enhancement combination until the target sharpness is greater than the preset sharpness threshold.

[0030] When the image clarity is greater than a preset clarity threshold, the multi-directional ceramic mesh enhanced image is used as the target multi-directional ceramic mesh image.

[0031] Optionally, constructing the wavelength matrix according to preset wavelength attributes includes:

[0032] Extract the wavelength range and wavelength color from the wavelength attributes;

[0033] The number of wavelengths is determined based on the wavelength range, wherein the formula for calculating the number of wavelengths is:

[0034]

[0035] Where D is the number of wavelengths, A1 is the end point in the wavelength range, A2 is the start point in the wavelength range, and L is the wavelength step size in the wavelength range;

[0036] The wavelength color is used as the row attribute of the wavelength matrix, and the wavelength range and the number of wavelengths are used as the column attributes of the wavelength matrix.

[0037] The wavelength matrix is ​​obtained by filling the row and column attributes with numerical values.

[0038] Optionally, generating the reflection matrix of the multi-directional ceramic mesh image based on the wavelength matrix includes:

[0039] The incident light rays are dispersed according to the wavelength range in the wavelength matrix to obtain the target incident dispersed light rays;

[0040] The intersection point of the rays in the multi-directional ceramic mesh image is determined based on the incident ray direction of the incident dispersed rays of the target.

[0041] The propagation distance of the light rays is determined based on the intersection point of the light rays and the starting point of the incident light ray direction;

[0042] The reflectivity of the multi-directional ceramic mesh image is calculated using the light propagation distance and the light intensity of the incident diffuse light from the target, wherein the reflectivity calculation formula is:

[0043]

[0044] Among them, H k Let I be the reflectance at the k-th wavelength color. ik Let S be the intensity of the incident diffuse light from the i-th target at the k-th wavelength color, α be the reflectance coefficient, and S be the intensity of the diffuse light from the i-th target. ik Let be the light propagation distance of the incident scattered light ray of the i-th target under the k-th wavelength color, and e be the number of incident scattered light rays of the target;

[0045] The reflection matrix of the multi-directional ceramic mesh image is generated based on the reflectivity.

[0046] Optionally, mapping the reflection matrix and the wavelength matrix to obtain the color spectrum matrix includes:

[0047] Extract the wavelength color attribute and reflectance attribute from the reflection matrix;

[0048] Extract the wavelength color attribute and wavelength range attribute from the wavelength matrix;

[0049] The mapping relationship between the reflection matrix and the wavelength matrix is ​​generated based on the wavelength color attributes;

[0050] The color spectrum matrix is ​​obtained by connecting the reflectivity attribute and the wavelength range attribute through the mapping relationship.

[0051] Optionally, generating the color spectral curve of the multi-directional ceramic grid image through the color spectral matrix includes:

[0052] The wavelength color in the color spectrum matrix is ​​used as the horizontal axis attribute;

[0053] Use the reflectance in the color spectral matrix as the vertical axis attribute;

[0054] The color spectrum curve of the multi-directional ceramic mesh image is generated based on the horizontal axis attribute and the vertical axis attribute.

[0055] Optionally, the step of performing mesh color aggregation on the multi-directional ceramic mesh image based on the correlation degree to generate the color distribution of the ceramic sample includes:

[0056] Generate a set of associated grids in the multi-directional ceramic grid image based on the correlation degree;

[0057] The target associated grids of the associated grid set are filtered according to the preset four directions;

[0058] The colors of the target associated mesh are aggregated to obtain the mesh aggregated color;

[0059] The target non-associated grid in the associated grid set is treated as a separate grid color;

[0060] The color distribution of the ceramic sample is generated based on the aggregated colors of the grid and the individual colors of the grid.

[0061] To address the above problems, the present invention also provides a ceramic color detection device, the device comprising:

[0062] The image sharpness calculation module is used to extract multi-angle ceramic images of a preset ceramic sample, divide the multi-angle ceramic images into grids to obtain multi-angle ceramic grid images, and calculate the image sharpness of the multi-angle ceramic grid images using a preset image sharpness algorithm.

[0063] The wavelength matrix construction module is used to perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes.

[0064] The color spectrum matrix generation module is used to generate a reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and to map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix.

[0065] The color spectrum curve generation module is used to generate the color spectrum curve of the multi-directional ceramic grid image through the color spectrum matrix, and extract the color spectrum features in the color spectrum curve.

[0066] The color distribution generation module is used to calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and the preset grid identifier, and to perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample.

[0067] This invention extracts grid features from multi-directional ceramic images to obtain information such as the shape, size, and arrangement of the ceramic grids, which is beneficial for subsequent image processing and analysis. By using an image sharpness algorithm to calculate the sharpness of the multi-directional ceramic grid images, the image clarity can be assessed, allowing for image enhancement processing to improve image quality and detail, making ceramic color detection more accurate and reliable. Constructing a wavelength matrix based on wavelength attributes provides foundational data for subsequent color analysis. Generating a reflection matrix from the multi-directional ceramic grid images based on the wavelength matrix reflects light reflection at different wavelengths. Mapping the reflection matrix to the wavelength matrix yields a color spectrum matrix, representing the color distribution of the multi-directional ceramic grid images. Analyzing the color spectrum matrix generates color spectrum curves and extracts color spectrum features, providing a more specific description of the ceramic sample's color. Calculating the correlation degree of the multi-directional ceramic grid images using grid identifiers assesses the color matching degree between different multi-directional ceramic grid images of the ceramic sample. Based on the correlation degree and color spectrum features, grid color aggregation is performed on the multi-directional ceramic grid images to generate the color distribution of the ceramic sample, providing more intuitive color information. Therefore, the ceramic color detection method and apparatus proposed in this invention can solve the problem of low accuracy in ceramic color detection. Attached Figure Description

[0068] Figure 1 This is a schematic flowchart of a ceramic color detection method provided in an embodiment of the present invention;

[0069] Figure 2 This is a schematic diagram of an image enhancement process provided in an embodiment of the present invention;

[0070] Figure 3 This is a schematic diagram of the process for generating a color spectral matrix according to an embodiment of the present invention;

[0071] Figure 4 This is a functional block diagram of a ceramic color detection device provided in an embodiment of the present invention;

[0072] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0073] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0074] This application provides a ceramic color detection method. The execution entity of the ceramic color detection method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the ceramic color detection method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0075] Reference Figure 1 The diagram shown is a schematic flowchart of a ceramic color detection method according to an embodiment of the present invention. In this embodiment, the ceramic color detection method includes:

[0076] S1. Extract multi-directional ceramic images of a preset ceramic sample, divide the multi-directional ceramic images into grids to obtain multi-directional ceramic grid images, and calculate the image clarity of the multi-directional ceramic grid images using a preset image clarity algorithm.

[0077] In this embodiment of the invention, the multi-angle ceramic image refers to the acquisition of images of the ceramic sample from multiple angles, such as an image of the ceramic sample acquired from a front view, an image of the ceramic sample acquired from a left view, and so on.

[0078] In this embodiment of the invention, the extraction of multi-directional ceramic images of a preset ceramic sample includes:

[0079] Generate multi-angle view attributes of ceramic samples based on preset view directions and preset view angles;

[0080] Generate multi-angle ceramic view states of ceramic samples one by one according to the multi-angle view attributes;

[0081] A multi-directional ceramic image of the ceramic sample is generated based on the multi-directional ceramic view state.

[0082] In detail, based on preset view directions and angles, multi-angle view attributes of ceramic samples can be generated at different angles and directions. For example, ceramic samples can be photographed or rendered from front, side, back, and other directions and at different angles (such as 0 degrees, 45 degrees, 90 degrees, etc.) to obtain multi-angle view attributes. Then, based on the multi-angle view attributes, multiple images from different directions and angles can be combined into one image to determine the state of the ceramic sample from different perspectives. Based on the multi-angle ceramic view states, multi-angle ceramic images of ceramic samples from different perspectives and states can be generated. For example, through rendering, stitching, or image processing techniques, ceramic images from multiple perspectives can be generated, thus providing more comprehensive and detailed information about the ceramic samples.

[0083] For example, when the view direction is frontal and the view angle is 180 degrees, a multi-view attribute of the ceramic sample can be obtained. Then, based on the view direction of frontal and the view angle of 180 degrees in the multi-view attribute, the view state of the ceramic sample is determined, that is, the image state presented by the ceramic sample under this view direction and view angle. This view state is then used as a ceramic image of the ceramic sample. Furthermore, ceramic images in different orientations can be determined based on different view directions and different view angles, that is, multi-view ceramic images, providing basic data for subsequent color processing and analysis of ceramic sample images.

[0084] Furthermore, it is necessary to divide the ceramic image in each direction of the multi-directional ceramic image into a grid to obtain a multi-directional ceramic grid image. That is, the multi-directional ceramic image is evenly divided into multiple small regions, each grid region corresponding to a small square or a region of a specific shape. Each grid region can be extracted separately to form a small ceramic grid image, or multiple grid regions can be merged together to form a large ceramic grid image, so that each grid can better capture the details and features of the ceramic sample to obtain the best visual effect and analysis results.

[0085] Furthermore, in order to continue ceramic color detection with the best ceramic image quality, it is necessary to objectively evaluate the multi-directional ceramic grid images. That is, to measure the clarity of the image by measuring the clarity, to help judge the quality of the image, and to select the image with the best clarity to show the details and features of the ceramic sample, so as to provide more accurate and reliable ceramic color detection results.

[0086] In this embodiment of the invention, the image clarity refers to the image clarity of the ceramic grid image corresponding to each direction in the multi-directional ceramic grid image. Based on the image clarity, image detail enhancement processing is performed on the ceramic image to improve the image quality and the accuracy of ceramic color detection.

[0087] In this embodiment of the invention, calculating the image sharpness of the multi-directional ceramic grid image using a preset image sharpness algorithm includes:

[0088] Extract the pixel attributes of the multi-directional ceramic mesh image;

[0089] The difference between each pixel in the multi-directional ceramic mesh image is calculated using a preset difference algorithm.

[0090] The image sharpness of the multi-directional ceramic mesh image is calculated using the following preset image sharpness algorithm based on the pixel attributes and the difference:

[0091]

[0092] Where Q is the image sharpness, M is the image row dimension in the pixel attribute, N is the image column dimension in the pixel attribute, and ΔF x Let ΔF be the difference between pixels (m,n) in the x-direction. y Let δ be the difference between pixels (m,n) in the y direction, and let δ be the sharpness optimization factor.

[0093] In detail, the pixel attributes include information such as the pixel grayscale value, position, and overall pixel dimension of each grid in the ceramic grid image of each orientation. The pixel attributes of the multi-orientation ceramic grid image can be obtained from a pre-stored storage area using computer statements with data-grabbing capabilities (such as Java statements, Python statements, etc.). It is also necessary to calculate the pixel difference corresponding to each grid in the ceramic grid image of each orientation. The pixel difference represents the difference between pixels in each grid in the x-direction and the y-direction. The difference reflects the degree of change of the pixel, i.e., the intensity of the edge or texture. The larger the difference value, the more obvious the change at that pixel. The difference algorithm includes, but is not limited to, the Sobel operator, the Prewitt operator, and the Roberts operator. For example, to calculate the difference of pixels in the x-direction for image F, ΔF... x (m,n) = F(m+1,n) - F(m,n), where F(m+1,n) represents the pixel value of image F at (m+1,n) and F(m,n) represents the pixel value of image F at (m,n). This represents the difference between the current pixel and its neighboring pixels, used to capture the changes in the image in the x-direction. This allows us to obtain the difference results of the entire image in the x-direction, and thus calculate the image sharpness based on pixel attributes and pixel differences. Image sharpness reflects the texture variation features and detail contrast of the image.

[0094] Specifically, based on the differences between different pixels in the x-direction and y-direction, the overall image's differences in the x-direction and y-direction can be determined. Based on the image's row dimension M and column dimension N, the overall image sharpness is calculated. The sharpness optimization factor δ is an adjustment parameter used to balance the effects of image sharpness and noise. This factor typically ranges from 0 to 1. When the sharpness optimization factor is 0, only edge details are considered, ignoring the influence of image noise, potentially resulting in a higher calculated image sharpness. When the sharpness optimization factor is 1, the overall image sharpness is considered, but details may be ignored, potentially resulting in a lower calculated image sharpness. Therefore, the value of the sharpness optimization factor δ can be customized according to the actual situation; generally, a value of 0.5 is used.

[0095] Furthermore, ceramic images can be processed based on their image clarity. If the image clarity is very low, image enhancement processing is required to reduce the impact of image blur, noise, and other factors on image clarity, making the image clearer, improving the visual effect of the image, and increasing the accuracy of ceramic color detection.

[0096] S2. Perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes.

[0097] In this embodiment of the invention, the target multi-directional ceramic mesh image refers to an image enhancement process performed on an image with low resolution to obtain a higher quality ceramic image corresponding to each direction.

[0098] In this embodiment of the invention, reference is made to Figure 2 As shown, the step of performing image enhancement processing on the multi-directional ceramic mesh image based on the image sharpness to obtain the target multi-directional ceramic mesh image includes:

[0099] S21. When the image sharpness is less than or equal to a preset sharpness threshold, a sharpness parameter enhancement combination is generated based on preset contrast parameters, preset sharpening parameters and preset resolution parameters.

[0100] S22. Perform image enhancement processing on the multi-directional ceramic mesh image according to the enhancement combination of the sharpness parameters to obtain a multi-directional ceramic mesh enhanced image;

[0101] S23. Calculate the target sharpness of the multi-directional ceramic mesh enhanced image. When the target sharpness is less than or equal to a preset sharpness threshold, adjust the sharpness parameter enhancement combination and return to the step of performing image enhancement processing on the multi-directional ceramic mesh image according to the sharpness parameter enhancement combination until the target sharpness is greater than the preset sharpness threshold.

[0102] S24. When the image clarity is greater than a preset clarity threshold, the multi-directional ceramic mesh enhanced image is used as the target multi-directional ceramic mesh image.

[0103] In detail, a sharpness threshold is preset as the basis for determining whether an image needs enhancement. If the image sharpness is less than or equal to the preset sharpness threshold, a sharpness parameter enhancement combination is generated based on preset contrast, sharpening, and resolution parameters. This sharpness parameter enhancement combination configures the contrast, sharpening, and resolution parameter values. Initially, an initial sharpness parameter enhancement combination is configured. The image contrast can be enhanced through methods such as histogram equalization, adaptive histogram equalization, or contrast stretching, resulting in a wider brightness range and more obvious details. Appropriate sharpening filters, such as Laplacian filters or high-pass filters, are used to enhance the image edges and details, making the image appear sharper. Alternatively, when the resolution of the target multi-directional ceramic mesh image is low, super-resolution reconstruction algorithms can be used to increase the image details and sharpness. Image processing techniques are used to upscale low-resolution images to high resolution. The parameters of each step of the image enhancement process can be adjusted and optimized based on the actual effect. The best image enhancement effect is achieved by repeatedly trying different parameter values.

[0104] Specifically, the image enhancement processing of the multi-directional ceramic mesh image is performed using the aforementioned sharpness parameter enhancement combination to obtain the enhanced image. That is, in ceramic images with sharpness less than a preset sharpness threshold, the ceramic image is optimized according to the contrast parameter, sharpening parameter value, and resolution parameter value in the sharpness parameter enhancement combination. Then, the target sharpness of the enhanced image is calculated. If the target sharpness is less than or equal to the preset sharpness threshold, the sharpness parameter enhancement combination is adjusted, and the group with the highest sharpness is selected as the sharpness parameter enhancement combination. The process is then returned to the image enhancement processing step and iterated until the target sharpness is greater than the preset sharpness threshold. If the image sharpness is greater than the preset sharpness threshold, the enhanced image is used as the target multi-directional ceramic mesh image.

[0105] Furthermore, after adjusting and optimizing the ceramic grid image corresponding to each orientation, color detection needs to be performed on the ceramic grid image of each orientation. That is, wavelength detection needs to be applied to the ceramic grid image. Therefore, a wavelength matrix needs to be constructed to realize color detection of the ceramic grid image.

[0106] In this embodiment of the invention, the wavelength matrix includes wavelength range and wavelength color correspondence. Then, according to the structured data form of the wavelength matrix, different color wavelengths are applied to each grid in the ceramic image in sequence, so that the color of each grid can be observed intuitively.

[0107] In this embodiment of the invention, constructing a wavelength matrix based on preset wavelength attributes includes:

[0108] Extract the wavelength range and wavelength color from the wavelength attributes;

[0109] The number of wavelengths is determined based on the wavelength range, wherein the formula for calculating the number of wavelengths is:

[0110]

[0111] Where D is the number of wavelengths, A1 is the end point in the wavelength range, A2 is the start point in the wavelength range, and L is the wavelength step size in the wavelength range;

[0112] The wavelength color is used as the row attribute of the wavelength matrix, and the wavelength range and the number of wavelengths are used as the column attributes of the wavelength matrix.

[0113] The wavelength matrix is ​​obtained by filling the row and column attributes with numerical values.

[0114] Specifically, the wavelength range includes the start and end points of the wavelength, as well as the wavelength step size, and thus determines the number of wavelengths; the wavelength color refers to the color corresponding to the wavelength of light. Within the visible spectrum, different wavelengths of light produce different color perceptions. Common wavelength colors include red, orange, yellow, green, cyan, blue, and violet. For example, the wavelength range for red is approximately 620-750 nanometers, for orange it is approximately 590-620 nanometers, for yellow it is approximately 570-590 nanometers, for green it is approximately 495-570 nanometers, for cyan it is approximately 450-495 nanometers, for blue it is approximately 450-495 nanometers, and for violet it is approximately 380-450 nanometers.

[0115] Specifically, wavelength attributes are used as the matrix dimensions of the wavelength matrix, wavelength color is used as the row attributes, and wavelength range is used as the column attributes. For example, if the wavelength colors are red, orange, yellow, green, cyan, blue, and purple, then red, orange, yellow, green, cyan, blue, and purple are used as the row attributes, and the start point, end point, step size, and number of wavelengths within the wavelength range are used as the column attributes. Then the wavelength matrix is ​​a 7×4 matrix, and the values ​​under each attribute are used to generate the wavelength matrix.

[0116] Furthermore, the wavelength matrix can be used to simulate the reflection and scattering of ceramic grids under real light, thereby understanding the optical properties of ceramic materials. This helps to more realistically showcase the appearance, color, and texture of ceramics, and improve the realism of the design.

[0117] S3. Generate the reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix.

[0118] In this embodiment of the invention, the reflection matrix refers to the reflectivity of each grid image in each orientation of the ceramic grid image. The reflection matrix constructed based on the grid properties and emissivity can simulate the reflection and scattering of the ceramic grid under real illumination.

[0119] In this embodiment of the invention, generating the reflection matrix of the multi-directional ceramic mesh image based on the wavelength matrix includes:

[0120] The incident light rays are dispersed according to the wavelength range in the wavelength matrix to obtain the target incident dispersed light rays;

[0121] The intersection point of the rays in the multi-directional ceramic mesh image is determined based on the incident ray direction of the incident dispersed rays of the target.

[0122] The propagation distance of the light rays is determined based on the intersection point of the light rays and the starting point of the incident light ray direction;

[0123] The reflectivity of the multi-directional ceramic mesh image is calculated using the light propagation distance and the light intensity of the incident diffuse light from the target, wherein the reflectivity calculation formula is:

[0124]

[0125] Among them, H k Let I be the reflectance at the k-th wavelength color. ik Let S be the intensity of the incident diffuse light from the i-th target at the k-th wavelength color, α be the reflectance coefficient, and S be the intensity of the diffuse light from the i-th target. ik Let be the light propagation distance of the incident scattered light ray of the i-th target under the k-th wavelength color, and e be the number of incident scattered light rays of the target;

[0126] The reflection matrix of the multi-directional ceramic mesh image is generated based on the reflectivity.

[0127] In detail, the incident light is dispersed according to the preset wavelength range to obtain the target incident dispersed light, that is, the incident light will be dispersed at different wavelengths to simulate the real spectrum. First, if the wavelength range is λ min To λ max The direction of the incident ray is d inFurthermore, the wavelength range is divided into several discrete wavelength points, which can be done using equal or non-equal intervals. Therefore, there are a total of N wavelength points, namely λ1, λ2, ..., λ... N For each wavelength point λ i According to the direction d of the incident light in Calculate the direction d of the incident diffuse ray from the target. dispersion This can be achieved by using the laws of refraction and dispersion of light to calculate the direction of the incident and dispersed rays of the target. This process is repeated until the directions of the incident and dispersed rays corresponding to all wavelengths are obtained, thus yielding the set of incident and dispersed rays, denoted as {d}. dispersion_1 ,d dispersion_2 ,...,d dispersion_N In the wavelength matrix, the wavelength range corresponding to each wavelength color is determined based on the number of wavelengths. If the number of wavelengths is 10, then there are 10 wavelength points. Therefore, there are 10 incident scattered rays corresponding to the red wavelength. Similarly, the incident scattered rays of other colors are also determined based on the number of wavelengths.

[0128] Specifically, by illuminating each grid of a ceramic mesh image from different orientations with different incident scattered rays, the intersection point between each incident scattered ray and the ceramic mesh can be determined, resulting in a set of ray intersection points denoted as {p1, p2, ..., p...}. N}, where each p i A coordinate point is a point on the ceramic grid where light rays intersect. These intersection points can be captured using optical tracking technology. Based on the intersection point and the starting point of the incident ray direction, the propagation distance of the light ray can be calculated using the Euclidean distance formula. Furthermore, the angle of incidence of the target's incident diffuse ray can be determined from the intersection point. This is achieved by measuring the angle at the intersection point, which equals the angle of reflection. Therefore, the angle of reflection can be determined from the angle of incidence, and the direction of the reflected ray can be generated from the angle of reflection, thus determining the reflected ray.

[0129] Furthermore, based on the light propagation distance and the intensity of the incident diffuse light from the target, the reflectivity of each grid in the ceramic grid image from each orientation can be calculated. For example, a multi-orientation ceramic grid image may include A, B, C, and D, where image A contains different grids, namely A1, A2, ..., A nBy illuminating A1 with light according to different wavelengths of color in the wavelength matrix, the intensity of incident light and reflected light of different wavelengths of color can be obtained, thus obtaining the reflectivity of the A1 grid under different wavelengths of color. The reflection coefficient α measures the reflection efficiency of light at the intersection point. The reflection coefficient is usually a real number between 0 and 1, representing the proportion of energy lost during reflection. The specific value of the reflection coefficient can be adjusted according to the material and surface characteristics. Then, the reflectivity on the grid is used to generate a reflection matrix. For example, the reflectivity of A1 under the red wavelength is H1, the reflectivity under the green wavelength is H2, and the reflectivity under the blue wavelength is H3, thus generating the reflection matrix of each grid under different wavelengths of color.

[0130] Furthermore, spectral information for different wavelengths of color can be obtained based on the wavelength matrix and reflection matrix. This is crucial for analyzing and understanding the color characteristics of light. Displaying the reflection characteristics of light in the form of a color image makes the results more intuitive and easier to understand, and can better convey the color information of light, thereby more accurately determining the color distribution of ceramic samples.

[0131] In this embodiment of the invention, the color spectral matrix is ​​generated by fusing the reflection matrix and the wavelength matrix, merging those with the same attributes, and thus representing the reflectivity distribution of light under different wavelength colors. It can be used to analyze and understand the optical properties of materials, as well as for visualization analysis and display.

[0132] In this embodiment of the invention, reference is made to Figure 3 As shown, the step of mapping the reflection matrix and the wavelength matrix to obtain the color spectrum matrix includes:

[0133] S31. Extract the wavelength color attribute and reflectance attribute from the reflection matrix;

[0134] S32. Extract the wavelength color attribute and wavelength range attribute from the wavelength matrix;

[0135] S33. Generate the mapping relationship between the reflection matrix and the wavelength matrix based on the wavelength color attribute;

[0136] S34. The reflectivity attribute and the wavelength range attribute are connected by the mapping relationship to obtain the color spectrum matrix.

[0137] In detail, the reflection matrix includes the correspondence between wavelength color and reflectivity, that is, the wavelength color attribute and reflectivity attribute are extracted from the reflection matrix; the wavelength matrix includes the correspondence between wavelength color and wavelength start value, wavelength end value, wavelength step size and step size number, that is, the wavelength color attribute and wavelength range attribute are extracted from the wavelength matrix. Since both the reflection matrix and the wavelength matrix have wavelength color attributes, the reflectivity attribute in the reflection matrix and the wavelength range attribute in the wavelength matrix are connected based on the wavelength color attribute to obtain the color spectrum matrix. In the color spectrum, the row attribute is the wavelength color attribute, and the column attributes are the wavelength start value, wavelength end value, wavelength step size, step size number and reflectivity.

[0138] Furthermore, the data in the color spectrum matrix is ​​displayed in the form of curves, making the color characteristics more intuitive and visible. By observing the color spectrum curves, we can understand the changes in reflectivity of light under different wavelengths of color, thereby gaining an intuitive perception of the color characteristics of ceramic mesh materials.

[0139] S4. Generate the color spectrum curve of the multi-directional ceramic grid image through the color spectrum matrix, and extract the color spectrum features in the color spectrum curve.

[0140] In this embodiment of the invention, the color spectrum curve mainly uses wavelength color and reflectance as the primary attributes, while other attributes in the color spectrum matrix are used as auxiliary attributes. When the reflectance has a large error, the reflectance can be recalculated through other attributes in the color spectrum matrix to ensure the accuracy of the reflectance.

[0141] In this embodiment of the invention, generating the color spectral curve of the multi-directional ceramic grid image through the color spectral matrix includes:

[0142] The wavelength color in the color spectrum matrix is ​​used as the horizontal axis attribute;

[0143] Use the reflectance in the color spectral matrix as the vertical axis attribute;

[0144] The color spectrum curve of the multi-directional ceramic mesh image is generated based on the horizontal axis attribute and the vertical axis attribute.

[0145] In detail, the wavelength color in the color spectrum matrix is ​​used as the horizontal axis attribute of the color spectrum curve, and the reflectance in the color spectrum matrix is ​​used as the vertical axis attribute of the color spectrum curve. That is, the color spectrum curve of the ceramic grid image in each direction is generated by the correspondence between wavelength color and reflectance. In other words, all points are connected in sequence to form a curve. This curve is the color spectrum curve of the multi-directional ceramic grid image, which visualizes the color information in the multi-directional ceramic grid image, thereby better analyzing and evaluating the color characteristics of ceramic products.

[0146] Specifically, a multi-view ceramic grid image may include a frontal ceramic image A, a left-side ceramic image B, a right-side ceramic image C, and a top-side ceramic image D. The frontal ceramic image A may contain different grids, namely A1, A2, ..., A... n The ceramic image B on the left also includes different grids, namely B1, B2, ..., B n The ceramic image C on the right also includes different grids, namely C1, C2, ..., C n Different wavelengths of color are applied to a grid, and the reflectance corresponding to different wavelengths of color can be applied to a grid. That is, each grid will correspond to a color spectrum curve.

[0147] Furthermore, based on the generated color spectrum curve, the peak and valley values ​​of reflectance corresponding to different wavelengths of color in the spectrum curve can be extracted. The peak value refers to the highest point or maximum value in the waveform, representing the maximum amplitude or maximum energy of the waveform; the valley value refers to the lowest point or minimum value in the waveform, representing the minimum amplitude or minimum energy of the waveform. For example, in grid A1, the reflectance corresponding to the red wavelength is the highest and the reflectance corresponding to the blue wavelength is the lowest. Therefore, the red wavelength corresponding to the peak value with the highest reflectance and the blue wavelength corresponding to the valley value with the lowest reflectance are extracted.

[0148] Furthermore, for the peak and valley values ​​corresponding to each grid, grids with the same color can be grouped together. Therefore, it is necessary to calculate the correlation between different grids and group them together to more clearly observe the color distribution of the ceramic sample.

[0149] S5. Calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and the preset grid identifier, and perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample.

[0150] In this embodiment of the invention, the correlation degree refers to the correlation degree of each grid. Different grids are merged together by the correlation degree to form a larger area, thereby obtaining a higher-level ceramic grid structure, which facilitates the analysis and evaluation of the color and quality of ceramic products.

[0151] In this embodiment of the invention, calculating the correlation degree of the multi-directional ceramic grid image based on the color spectral features and preset grid identifiers includes:

[0152] Extract the reflectance peak from the color spectral features;

[0153] The grid region identifier of each grid in the multi-directional ceramic grid image is determined according to the preset grid identifier;

[0154] The correlation degree of the multi-directional ceramic grid image is calculated based on the peak reflectance and the grid region identifier, wherein the correlation degree calculation formula is:

[0155]

[0156] Where G is the correlation degree, u t To identify the peak reflectance value corresponding to the t-th grid region, ρ t1 To identify the wavelength color corresponding to the reflectivity peak of the t-th grid region, u p The p-th grid region is identified by its corresponding reflectance peak value, ρ. p2 Identify the wavelength color corresponding to the reflectivity peak of the p-th grid region.

[0157] In detail, the color spectral features include reflectance peaks and reflectance valleys. Based on the reflectance peaks, the correlation between grids can be determined. The reflectance peaks in the color spectral features can be extracted using computer statements with data capture capabilities. In addition, the grid region identifier of each grid in the multi-directional ceramic grid image can be determined according to the grid identifier. For example, the grid size corresponding to the frontal ceramic image A is 4×4, with a total of 16 grid regions. Each grid region corresponds to a unique identifier, such as the first grid region corresponding to A1, the second grid region corresponding to A2, and so on, to obtain the region identifiers corresponding to all grid regions. Then, each grid region can be accessed through an index.

[0158] Specifically, the wavelength colors corresponding to the reflectance peaks in different grids are compared one by one. If the wavelength colors corresponding to the reflectance peaks are the same, it indicates that the two grids have a certain correlation; otherwise, they do not have a correlation. t →ρ t1 The wavelength color corresponding to the reflectance peak of the t-th grid region is ρ. t1 u p →ρ p2 The wavelength color corresponding to the reflectance peak of the p-th grid region is ρ. p2 If ρ t1 With ρ p2 If they are the same, it means that the two grid regions are related.

[0159] For example, if the grid size corresponding to the front-facing ceramic image A is 2×2, then it has four grid area labels: A1, A2, A3, and A4. The wavelength colors corresponding to the reflectance peaks in the four grid area labels A1, A2, A3, and A4 are compared. If the wavelength colors of A1 and A2 are the same, it means that the grid areas corresponding to A1 and A2 are related. If the wavelength colors of A2 and A3 are the same, then the grid areas corresponding to A2 and A3 are not related. That is, if a peak is observed in the red light wavelength range and a valley is observed in the green light wavelength range, it can be inferred that the area has a red or near-red color.

[0160] Furthermore, by aggregating the colors of adjacent grids, the degree of grid color aggregation can be controlled, thereby achieving different color distribution effects, enhancing the visual effect of ceramic samples, and improving the accuracy of color detection of ceramic samples.

[0161] In this embodiment of the invention, the color distribution refers to the distribution of different colors in ceramic images from different orientations, thereby determining the overall color of the ceramic sample.

[0162] In this embodiment of the invention, the step of performing grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample includes:

[0163] Generate a set of associated grids in the multi-directional ceramic grid image based on the correlation degree;

[0164] The target associated grids of the associated grid set are filtered according to the preset four directions;

[0165] The colors of the target associated mesh are aggregated to obtain the mesh aggregated color;

[0166] The target non-associated grid in the associated grid set is treated as a separate grid color;

[0167] The color distribution of the ceramic sample is generated based on the aggregated colors of the grid and the individual colors of the grid.

[0168] In detail, grids with the same color in the image are grouped together, i.e., associated grid sets. In the associated grid sets, target associated grids are filtered by four directions. The target associated grids are grids connected in four directions. The colors of the target associated grids are aggregated to form a larger color area. Non-target associated grids in the associated grid set are regarded as individual colors of the grids. Based on the aggregated grid colors and individual grid colors, the color distribution of the ceramic sample is generated, thereby determining the overall color distribution of the ceramic sample.

[0169] For example, if the grid size corresponding to the frontal ceramic image A is 3×3, and A1 and A2 are related, A1 and A3 are related, and A1 and A5 are related, then the set of related grids generated by the grids associated with A1 is {A2, A3, A5}. By filtering the target related grids in the four directions of A1 (top, bottom, left, and right), there are no grids above A1, A4 below A1, no grids to the left of A1, and A2 to the right of A1. Therefore, the target related grid of A1 is A2. Then, A1 and A2 are merged into a larger grid area to obtain the grid aggregated color, while A3 and A5 are treated as separate colors. Their separate grids can be merged with the next central grid to obtain different grid aggregated colors. The color distribution of the ceramic sample is generated based on the grid aggregated colors and the individual grid colors.

[0170] This invention extracts grid features from multi-directional ceramic images to obtain information such as the shape, size, and arrangement of the ceramic grids, which is beneficial for subsequent image processing and analysis. By using an image sharpness algorithm to calculate the sharpness of the multi-directional ceramic grid images, the image clarity can be assessed, allowing for image enhancement processing to improve image quality and detail, making ceramic color detection more accurate and reliable. Constructing a wavelength matrix based on wavelength attributes provides foundational data for subsequent color analysis. Generating a reflection matrix from the multi-directional ceramic grid images based on the wavelength matrix reflects light reflection at different wavelengths. Mapping the reflection matrix to the wavelength matrix yields a color spectrum matrix, representing the color distribution of the multi-directional ceramic grid images. Analyzing the color spectrum matrix generates color spectrum curves and extracts color spectrum features, providing a more specific description of the ceramic sample's color. Calculating the correlation degree of the multi-directional ceramic grid images using grid identifiers assesses the color matching degree between different multi-directional ceramic grid images of the ceramic sample. Based on the correlation degree and color spectrum features, grid color aggregation is performed on the multi-directional ceramic grid images to generate the color distribution of the ceramic sample, providing more intuitive color information. Therefore, the ceramic color detection method and apparatus proposed in this invention can solve the problem of low accuracy in ceramic color detection.

[0171] like Figure 4 The diagram shown is a functional block diagram of a ceramic color detection device provided in an embodiment of the present invention.

[0172] The ceramic color detection device 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the ceramic color detection device 100 may include an image sharpness calculation module 101, a wavelength matrix construction module 102, a color spectrum matrix generation module 103, a color spectrum curve generation module 104, and a color distribution generation module 105. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0173] In this embodiment, the functions of each module / unit are as follows:

[0174] The image sharpness calculation module 101 is used to extract multi-angle ceramic images of a preset ceramic sample, divide the multi-angle ceramic images into grids to obtain multi-angle ceramic grid images, and calculate the image sharpness of the multi-angle ceramic grid images using a preset image sharpness algorithm.

[0175] The wavelength matrix construction module 102 is used to perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes.

[0176] The color spectrum matrix generation module 103 is used to generate a reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and to map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix.

[0177] The color spectrum curve generation module 104 is used to generate the color spectrum curve of the multi-directional ceramic grid image through the color spectrum matrix and extract the color spectrum features in the color spectrum curve.

[0178] The color distribution generation module 105 is used to calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and the preset grid identifier, and to perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree and the color spectral characteristics to generate the color distribution of the ceramic sample.

[0179] In detail, each module in the ceramic color detection device 100 described in this embodiment of the invention adopts the same characteristics as described above during use. Figures 1 to 3 The method used is the same as the ceramic color detection method described above and can produce the same technical effect, so it will not be repeated here.

[0180] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0181] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0182] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0183] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0184] Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects. The scope of the invention is not limited to the foregoing description, and all variations within the meaning and scope of equivalents falling within the protection scope are intended to be included in the invention.

[0185] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0186] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices described in the system embodiments may also be implemented by a single unit or device through software or hardware. Terms such as "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting the color of ceramics, characterized in that, The method includes: S1. Extract multi-directional ceramic images of a preset ceramic sample, divide the multi-directional ceramic images into grids to obtain multi-directional ceramic grid images, and calculate the image clarity of the multi-directional ceramic grid images using a preset image clarity algorithm. S2. Perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes; S3. Generate the reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix; S4. Generate the color spectral curve of the multi-directional ceramic grid image through the color spectral matrix, and extract the color spectral features in the color spectral curve; S5. Calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers; perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample; wherein calculating the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers includes: S51. Extract the reflectance peak value from the color spectral features; S52. Determine the grid region identifier of each grid in the multi-directional ceramic grid image according to the preset grid identifier; S53. Calculate the correlation degree of the multi-directional ceramic grid image based on the peak reflectance and the grid region identifier, wherein the correlation degree calculation formula is: Where G is the correlation degree, u t To identify the peak reflectance value corresponding to the t-th grid region, ρ t1 To identify the wavelength color corresponding to the reflectivity peak of the t-th grid region, u p The p-th grid region is identified by its corresponding reflectance peak value, ρ. p2 Identify the wavelength color corresponding to the reflectivity peak of the p-th grid region.

2. The ceramic color detection method as described in claim 1, characterized in that, The extraction of multi-directional ceramic images from a preset ceramic sample includes: Generate multi-angle view attributes of ceramic samples based on preset view directions and preset view angles; Generate multi-angle ceramic view states of ceramic samples one by one according to the multi-angle view attributes; A multi-directional ceramic image of the ceramic sample is generated based on the multi-directional ceramic view state.

3. The ceramic color detection method as described in claim 1, characterized in that, The step of calculating the image sharpness of the multi-directional ceramic grid image using a preset image sharpness algorithm includes: Extract the pixel attributes of the multi-directional ceramic mesh image; The difference between each pixel in the multi-directional ceramic mesh image is calculated using a preset difference algorithm. The image sharpness of the multi-directional ceramic mesh image is calculated using the following preset image sharpness algorithm based on the pixel attributes and the difference: Where Q is the image sharpness, M is the image row dimension in the pixel attribute, N is the image column dimension in the pixel attribute, and ΔF x Let ΔF be the difference between pixels (m,n) in the x-direction. y Let δ be the difference between pixels (m,n) in the y direction, and let δ be the sharpness optimization factor.

4. The ceramic color detection method as described in claim 1, characterized in that, The step of performing image enhancement processing on the multi-directional ceramic mesh image based on the image clarity to obtain the target multi-directional ceramic mesh image includes: When the image sharpness is less than or equal to a preset sharpness threshold, a sharpness parameter enhancement combination is generated based on preset contrast parameters, preset sharpening parameters, and preset resolution parameters. The multi-directional ceramic mesh image is enhanced according to the aforementioned sharpness parameter enhancement combination to obtain a multi-directional ceramic mesh enhanced image; Calculate the target sharpness of the multi-directional ceramic mesh enhanced image. When the target sharpness is less than or equal to a preset sharpness threshold, adjust the sharpness parameter enhancement combination and return to the step of performing image enhancement processing on the multi-directional ceramic mesh image according to the sharpness parameter enhancement combination until the target sharpness is greater than the preset sharpness threshold. When the image clarity is greater than a preset clarity threshold, the multi-directional ceramic mesh enhanced image is used as the target multi-directional ceramic mesh image.

5. The ceramic color detection method as described in claim 1, characterized in that, The step of constructing a wavelength matrix based on preset wavelength attributes includes: Extract the wavelength range and wavelength color from the wavelength attributes; The number of wavelengths is determined based on the wavelength range, wherein the formula for calculating the number of wavelengths is: Where D is the number of wavelengths, A1 is the end point in the wavelength range, A2 is the start point in the wavelength range, and L is the wavelength step size in the wavelength range; The wavelength color is used as the row attribute of the wavelength matrix, and the wavelength range and the number of wavelengths are used as the column attributes of the wavelength matrix. The wavelength matrix is ​​obtained by filling the row and column attributes with numerical values.

6. The ceramic color detection method as described in claim 1, characterized in that, The step of generating the reflection matrix of the multi-directional ceramic mesh image based on the wavelength matrix includes: The incident light rays are dispersed according to the wavelength range in the wavelength matrix to obtain the target incident dispersed light rays; The intersection point of the rays in the multi-directional ceramic mesh image is determined based on the incident ray direction of the incident dispersed rays of the target. The propagation distance of the light rays is determined based on the intersection point of the light rays and the starting point of the incident light ray direction; The reflectivity of the multi-directional ceramic mesh image is calculated using the light propagation distance and the light intensity of the incident diffuse light from the target, wherein the reflectivity calculation formula is: Among them, H k Let I be the reflectance at the k-th wavelength color. ik Let S be the intensity of the incident diffuse light from the i-th target at the k-th wavelength color, α be the reflectance coefficient, and S be the intensity of the diffuse light from the i-th target. ik Let be the light propagation distance of the incident scattered light ray of the i-th target under the k-th wavelength color, and e be the number of incident scattered light rays of the target; The reflection matrix of the multi-directional ceramic mesh image is generated based on the reflectivity.

7. The ceramic color detection method as described in claim 1, characterized in that, The step of mapping the reflection matrix and the wavelength matrix to obtain the color spectrum matrix includes: Extract the wavelength color attribute and reflectance attribute from the reflection matrix; Extract the wavelength color attribute and wavelength range attribute from the wavelength matrix; The mapping relationship between the reflection matrix and the wavelength matrix is ​​generated based on the wavelength color attributes; The color spectrum matrix is ​​obtained by connecting the reflectivity attribute and the wavelength range attribute through the mapping relationship.

8. The ceramic color detection method as described in claim 1, characterized in that, The process of generating the color spectral curve of the multi-directional ceramic mesh image using the color spectral matrix includes: The wavelength color in the color spectrum matrix is ​​used as the horizontal axis attribute; Use the reflectance in the color spectral matrix as the vertical axis attribute; The color spectrum curve of the multi-directional ceramic mesh image is generated based on the horizontal axis attribute and the vertical axis attribute.

9. The ceramic color detection method as described in claim 1, characterized in that, The step of performing grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample includes: Generate a set of associated grids in the multi-directional ceramic grid image based on the correlation degree; The target associated grids of the associated grid set are filtered according to the preset four directions; The colors of the target associated mesh are aggregated to obtain the mesh aggregated color; The target non-associated grid in the associated grid set is treated as a separate grid color; The color distribution of the ceramic sample is generated based on the aggregated colors of the grid and the individual colors of the grid.

10. A ceramic color detection device, characterized in that, The device includes: The image sharpness calculation module is used to extract multi-angle ceramic images of a preset ceramic sample, divide the multi-angle ceramic images into grids to obtain multi-angle ceramic grid images, and calculate the image sharpness of the multi-angle ceramic grid images using a preset image sharpness algorithm. The wavelength matrix construction module is used to perform image enhancement processing on the multi-directional ceramic mesh image according to the image clarity to obtain the target multi-directional ceramic mesh image, and construct a wavelength matrix according to the preset wavelength attributes. The color spectrum matrix generation module is used to generate a reflection matrix of the multi-directional ceramic grid image based on the wavelength matrix, and to map the reflection matrix and the wavelength matrix to obtain the color spectrum matrix. The color spectrum curve generation module is used to generate the color spectrum curve of the multi-directional ceramic grid image through the color spectrum matrix, and extract the color spectrum features in the color spectrum curve. A color distribution generation module is used to calculate the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers, and to perform grid color aggregation on the multi-directional ceramic grid image based on the correlation degree to generate the color distribution of the ceramic sample. The step of calculating the correlation degree of the multi-directional ceramic grid image based on the color spectral characteristics and preset grid identifiers includes: Extract the reflectance peak from the color spectral features; determine the grid region identifier of each grid in the multi-directional ceramic grid image according to the preset grid identifier; calculate the correlation degree of the multi-directional ceramic grid image based on the reflectance peak and the grid region identifier, wherein the correlation degree calculation formula is: Where G is the correlation degree, u t To identify the peak reflectance value corresponding to the t-th grid region, ρ t1 To identify the wavelength color corresponding to the reflectivity peak of the t-th grid region, u p The p-th grid region is identified by its corresponding reflectance peak value, ρ. p2 Identify the wavelength color corresponding to the reflectivity peak of the p-th grid region.

Images