Intelligent Interaction Method and System Based on Inkjet Printing Color Recommendation

Abstract

This invention relates to an intelligent interactive method and system for inkjet printing color recommendation, belonging to the field of color management technology. The method includes: acquiring brightness value data of fabric images and printing template images; quantifying the gloss of the fabric surface based on the changes in brightness values ​​between different fabric images; quantifying the texture complexity of the fabric surface based on the distribution of brightness values ​​between pixels in each fabric image; correcting the color data of the printing template image to obtain corrected target color data; calculating the theoretical spectral reflectance corresponding to different ratios of mixed inks to obtain predicted color data of the mixed inks; matching the predicted color with the smallest color difference from the target color to obtain the corresponding optimal ink ratio parameters. This invention, by quantifying the gloss and texture complexity of the fabric, corrects traditional color prediction models, which can solve the deviation problem between the actual color and the screen color, and achieve high-precision prediction and recommendation of inkjet colors.

Description

Technical Field

[0001] This invention relates to the field of color management technology, and in particular to an intelligent interactive method and system for recommending colors based on inkjet printing. Background Technology

[0002] Inkjet printing technology boasts advantages such as high pattern precision and a wide range of color choices, enabling personalized customization to meet diverse market demands. Furthermore, unlike traditional plate-based printing, it employs contactless printing, minimizing fabric damage, increasing production efficiency, and being relatively environmentally friendly, thus gaining widespread application in the textile industry. However, a long-standing technological challenge remains: ensuring that the final color printed on the physical fabric closely matches the color seen by the designer on the digital screen. This "what you see is what you get" color fidelity directly impacts product quality and the realization of the designer's intentions.

[0003] To address this issue, the industry has proposed a series of color management and prediction technologies, among which the Kubelka-Monk-based color mixing prediction algorithm, namely... This theory, due to its solid physical optics foundation, has attracted considerable attention. It predicts the color formed on fabric by a specific ink ratio by establishing the absorption coefficient K and scattering coefficient S of the ink and fabric. This method explains the cause of color formation from a physical mechanism perspective and has better predictive accuracy than simple empirical color lookup tables.

[0004] However, traditional The theory has a fundamental limitation: its model assumes that light undergoes ideal diffuse reflection in a medium, meaning it is scattered uniformly in all directions. This assumption limits the model to essentially describing only the color of a perfectly matte surface. However, in real-world applications, many non-matte fabrics exist, such as silk and satin with a strong sheen, or cotton, linen, and canvas with a rough, textured feel. Their surface optical properties play a crucial role in the final visual color perception. Gloss creates highlights, and texture forms microscopic shadows; these complex physical phenomena go beyond traditional methods. The theory only has two parameters to describe it: the absorption coefficient K and the scattering coefficient S. Therefore, when directly applying traditional... When the theory predicts the color of these fabrics with complex surface characteristics, it cannot quantify and compensate for the visual effects of factors such as gloss and texture, resulting in a significant deviation between the predicted results and the colors actually observed by the human eye. This severely restricts its application in the field of high-end textile printing. Summary of the Invention

[0005] To achieve high-precision prediction and recommendation of inkjet colors, overcome the limitation of traditional KM theory which only applies to matte surfaces, and address the significant discrepancy between actual and screen colors when predicting colors for fabrics with complex surface characteristics, this invention provides an intelligent interactive method and system for inkjet printing color recommendation. The technical solution is as follows:

[0006] In a first aspect, the present invention provides an intelligent interactive method for color recommendation based on inkjet printing, comprising the following steps: acquiring fabric images and printing template images taken from multiple angles and converting them into Lab color space format to obtain brightness value data for each image; quantifying the gloss of the fabric surface based on the changes in brightness values ​​between the fabric images; quantifying the texture complexity of the fabric surface based on the distribution of brightness values ​​between pixels in each fabric image; correcting the color data of the printing template image based on the gloss and texture complexity of the fabric surface to obtain corrected target color data; calculating the theoretical spectral reflectance corresponding to different ratios of mixed inks to obtain predicted color data of the mixed inks; and matching the predicted color with the smallest color difference between the corrected printing template image and the target color to obtain the corresponding optimal ink ratio parameters.

[0007] Specifically, by compensating for the diffuse reflection attenuation of brightness visual effect caused by the complexity of fabric texture, suppressing the highlight effect caused by fabric gloss, and adjusting the brightness component of the printing template image, brightness correction of the printing template image is achieved; by reducing the negative impact of visual mixing and mutual interference of color light caused by the complexity of fabric texture on the color saturation presentation effect, suppressing the influence of reflected light caused by fabric gloss on the saturation visual effect, and adjusting the color component of the printing template image, saturation correction of the printing template image is achieved.

[0008] Preferably, a fabric sample to be printed is prepared, a standard light source is arranged to simulate natural light with the light source angle perpendicular to the fabric, and under the illumination of the standard light source, the camera lens is positioned around the fabric to take pictures of the fabric from multiple angles and directions. The shooting angle is determined by the angle between the camera lens and the normal of the fabric surface, and images of the fabric at different angles in the unprinted state are collected; information data of the printing template image is obtained, and the multi-angle fabric images and printing template images are converted into Lab color space format to obtain the brightness value data of all pixels in each image.

[0009] Preferably, based on the brightness value data of each fabric image, the number of different angles used when shooting the fabric images is obtained, the average brightness value of all fabric images under each shooting angle is calculated, then the mean of the brightness values ​​of all fabric images under all shooting angles, and the variance of the average brightness values ​​under all shooting angles are calculated. A logarithmic transformation is used to compress the variance and enhance the non-linear characteristics of the brightness values ​​between fabric images, thus obtaining the gloss characteristic value of the fabric surface. The function normalizes the gloss characteristic values ​​and uses the normalized values ​​as the gloss of the fabric surface.

[0010] Preferably, based on the brightness value data of each fabric image, the total number of fabric images, the number of pixels in each fabric image, and the brightness value of each pixel are obtained. The average brightness value of each fabric image is calculated, and the brightness value range of 1 to 100 is divided into multiple equal-length intervals. Based on the brightness value, the pixels are divided into each interval. The number of pixels in each interval on each fabric image is counted, and the proportion of the number of pixels in each interval on the corresponding fabric image is calculated.

[0011] Preferably, the variance and information entropy of the pixel brightness values ​​in each fabric image are calculated. The sum of the variance and information entropy for each fabric image is used as the brightness distribution feature value of the corresponding fabric image. The mean of the brightness distribution feature values ​​of all fabric images is calculated. The function normalizes the mean of the brightness distribution characteristic values ​​and uses the normalized value as the texture complexity of the fabric surface.

[0012] Preferably, the gloss and texture complexity of the fabric surface are obtained, the brightness data of the printing template image is extracted as the base brightness, and the diffuse reflection suppression coefficient and specular highlight correction coefficient are set according to the material characteristics of the fabric in the actual application scenario. The product between texture complexity and diffuse reflection suppression coefficient is used as brightness supplement coefficient, the product between gloss and specular highlight correction coefficient is used as brightness suppression coefficient, the difference between brightness supplement coefficient and brightness suppression coefficient is used as brightness correction coefficient, the product between brightness correction coefficient and base brightness is used as brightness correction value, and the sum of brightness correction value and base brightness is used as the corrected target color data.

[0013] Preferably, the gloss and texture complexity of the fabric surface are obtained, and the color component data of the printing template image are extracted as the basic color components. According to the material characteristics of the fabric and the application environment of the printed fabric in the actual application scenario, the specular reflection saturation enhancement coefficient and the diffuse reflection saturation enhancement coefficient are set. The product between gloss and specular reflection saturation enhancement coefficient is used as the saturation enhancement coefficient, the product between texture complexity and diffuse reflection saturation enhancement coefficient is used as the saturation supplement coefficient, the sum of saturation enhancement coefficient and saturation supplement coefficient is used as the saturation correction coefficient, the product between saturation correction coefficient and basic color components is used as the saturation correction value, and the sum of saturation correction value and basic color components is used as the color component value in the corrected target color data.

[0014] Preferred, according to Theoretical calculations were performed to obtain the theoretical spectral reflectance of different ink mixture ratios on an ideal fabric, and the results were then used to find the appropriate standard light source model. Standard documentation provides the standard observer function, the light source spectral power distribution function, and the reference white point. The theoretical spectral reflectance is converted to a value based on the standard observer function and the spectral power distribution function of the light source. Tristimulus values, then referenced by white points value pairs The tristimulus values ​​are normalized to obtain the predicted color values ​​of the mixed inks with the corresponding ratios on the ideal fabric. The predicted color data of the mixed inks with different ratios on the ideal fabric are then calculated sequentially.

[0015] Preferably, adopt The function constructs a color difference minimization objective function or a global search algorithm to find the predicted color with the smallest color difference from the target color as the final recommended color. Based on the mapping relationship between ink ratio and predicted color, the ink ratio parameter corresponding to the predicted color with the smallest color difference is obtained and used as the optimal ink ratio parameter. The optimal ink ratio parameter is used as the execution printing parameter of the inkjet printer and output to the inkjet printer for printing.

[0016] Secondly, the present invention provides an intelligent interactive system based on inkjet printing color recommendation for implementing the above-mentioned intelligent interactive method. The system includes: an inkjet printing device, a processor, a memory, a communication interface, a standard light source, and an image acquisition device. The processor stores computer program instructions for implementing the above-mentioned intelligent interactive method. The image acquisition device is capable of multi-directional and multi-angle adjustment. The standard light source can simulate the illumination environment of natural light. The communication interface is communicatively connected to the image acquisition device and the inkjet printing device.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] This invention achieves a quantitative assessment of fabric surface gloss and texture complexity by acquiring and analyzing fabric images from multiple angles. Compared to traditional single-angle detection methods, this method can comprehensively capture changes in the optical properties of the fabric under different viewing angles, effectively improving the robustness of the color correction model. Simultaneously, by constructing gloss and texture complexity, the surface physical and optical properties of the fabric are quantified and incorporated into the color prediction process. This compensates for visual deviations caused by factors such as gloss and texture. Especially for fabrics with complex, non-ideal surfaces such as silk, satin, and linen, where traditional methods struggle with color prediction, this intelligent interactive method significantly improves the accuracy of color prediction. This results in a higher degree of matching between the final printed color and the color displayed on the screen of the digitally printed image, thus enabling high-precision prediction and recommendation of inkjet colors for fabrics with complex surface properties. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the implementation of this intelligent interaction method.

[0020] Figure 2 This is a structural block diagram of the intelligent interactive system; Detailed Implementation

[0021] The technical features of the present invention will be further described in detail below with reference to the accompanying drawings, so that those skilled in the art can understand them.

[0022] An intelligent interactive method based on inkjet printing color recommendation, the implementation process of which is as follows: Figure 1 As shown, the specific implementation steps are as follows:

[0023] Step S1: Acquire fabric images and printing template images taken from multiple angles and convert them to Lab color space format to obtain the brightness value data of each image;

[0024] Specifically, prepare a fabric sample to be printed, simulate natural light by setting up a standard light source with the light source angle perpendicular to the fabric, and under the illumination of the standard light source, take multi-directional and multi-angle shots of the fabric from all sides. The shooting angle is determined by the angle between the camera lens and the normal of the fabric surface, and images of the fabric at different angles in the unprinted state are collected. Obtain information data of the printing template image, and convert both the multi-angle fabric images and the printing template images into Lab color space format to obtain the brightness value data of all pixels in each image. The selection of the standard light source and shooting angle can be adjusted according to the actual application scenario. Preferably, a standard light source of model D65 is selected for shooting. Shooting angles with the angle between the camera lens and the normal of the fabric surface of 0°, 30°, 45°, and 60° can be used to shoot the fabric sample. For each shooting angle, fabric images from the front, back, left, and right sides of the fabric sample should be collected.

[0025] In addition, generally speaking, since the initially acquired printing template image and the fabric image captured by the camera are... Data in format, This is a color space used for screen display, and its color representation may differ across different devices; while The format is a color space used to simulate the human eye's perception of color; it is device-independent and its presentation is unaffected by the device. The human eye is based on... Colors confirmed in a color space may exhibit color deviation when directly used for inkjet printing; therefore, by... Color data in the format converted to The color space format can map "screen perception" to "actual human eye perception", providing a unified perceptual benchmark for adjusting ink ratios during subsequent inkjet printing.

[0026] Step S2: Quantify the gloss of the fabric surface based on the changes in brightness values ​​between different fabric images;

[0027] The gloss of a fabric is essentially the ratio of specular reflection to diffuse reflection on its surface. Higher gloss indicates stronger specular reflection, leading to more drastic changes in the brightness of individual pixels at different angles. This dilutes the color effect created by diffuse reflection, resulting in increased perceived brightness and decreased color saturation. The gloss characteristics of a fabric can be reflected by observing the brightness variations between images from multiple angles. Higher gloss results in more dramatic changes in brightness values ​​at different angles, leading to a greater variance in brightness values ​​between images from different angles. By analyzing this variance, the gloss of the fabric can be quantified, allowing for adjustments to the impact of gloss on color brightness and saturation in subsequent steps.

[0028] Specifically, based on the brightness value data of each fabric image, the number of different angles used when capturing the fabric images is obtained. The average brightness value of all fabric images under each shooting angle is calculated. Then, the mean of the brightness values ​​of all fabric images under all shooting angles, as well as the variance of the average brightness values ​​under all shooting angles, are calculated. A logarithmic transformation is used to compress the variance and enhance the non-linear characteristics of the brightness values ​​between fabric images, thus obtaining the gloss characteristic value of the fabric surface. The function normalizes the gloss characteristic values ​​and uses the normalized gloss characteristic values ​​as the gloss of the fabric surface.

[0029] The process and specific formula for calculating gloss based on the difference in brightness values ​​between fabric images from different angles are as follows:

[0030]

[0031]

[0032]

[0033]

[0034] In the formula, Indicates the number of shooting angles. Indicates the first The average brightness value of fabric images from various shooting angles. This represents the average brightness value of all fabric images taken from all shooting angles. This represents the variance of the average brightness value among fabric images taken from all angles. This represents the characteristic value of the fabric surface's gloss. Indicates the gloss of the fabric surface; variance The greater the gloss of the fabric surface The higher the variance, the better. Too large an amount will reduce gloss. The value of the target color data is limited to a reasonable range by using logarithmic transformation to compress and restrict it, and normalization is performed on the gloss characteristic value. This is to ensure that the target color data value is within a reasonable range during the subsequent correction of the printing template image.

[0035] Step S3: Based on the distribution of brightness values ​​between pixels in each fabric image, quantify the texture complexity of the fabric surface;

[0036] The texture of a fabric is essentially its microscopic surface morphology. The more complex the texture, the rougher the surface, and the higher the degree of unevenness and disorder in its microscopic morphology. This leads to increased randomness in microscopic scattering, reducing the perceived brightness and saturation of colors. The distribution of brightness values ​​among pixels in a fabric image reflects the texture characteristics of the fabric surface; the more complex the texture, the more disordered the brightness distribution of pixels in the fabric image. By calculating and statistically analyzing the variance and information entropy of the brightness values ​​of pixels in various fabric images, where variance quantifies the amplitude of brightness fluctuations and information entropy quantifies the disorder of brightness distribution, the combined effect comprehensively describes the texture complexity of the fabric from two complementary dimensions: "range of fluctuation" and "degree of disorder." This aligns with the physical mechanisms of visual perception and optical scattering, thus quantifying the texture complexity of the fabric surface to facilitate correction of its impact on color brightness and saturation in subsequent steps.

[0037] Specifically, based on the brightness value data of each fabric image, the total number of fabric images, the number of pixels in each fabric image, and the brightness value of each pixel are obtained. The average brightness value of each fabric image is calculated, and the brightness value range of 1 to 100 is divided into multiple equal-length intervals. Based on the brightness value, the pixels are divided into each interval, and the number of pixels in each interval on each fabric image is counted. The percentage of pixels in each interval on the corresponding fabric image is also calculated. The number of intervals can be adjusted according to the actual application scenario. Typically, the number of intervals ranges from 16 to 64. Preferably, the brightness value range can be divided into 64 equal-length intervals.

[0038] Furthermore, the variance and information entropy of the pixel brightness values ​​in each fabric image are calculated. The sum of the variance and information entropy for each fabric image is used as the brightness distribution characteristic value of that fabric image. The mean of the brightness distribution characteristic values ​​of all fabric images is then calculated. The function normalizes the mean of the brightness distribution characteristic values ​​and uses the normalized value as the texture complexity of the fabric surface.

[0039] The process and specific formula for calculating gloss, based on the distribution characteristics of brightness values ​​among pixels in each fabric image, are as follows:

[0040]

[0041]

[0042]

[0043]

[0044]

[0045] In the formula, This indicates the total number of fabric images. Indicates the first The number of pixels in a fabric image. Indicates the first The first fabric image The brightness value of each pixel Indicates the first The average brightness value of the fabric images, where K represents the number of intervals. Indicates that the brightness value is located at the th The number of pixels in each interval in the th... The proportion of each fabric image; Indicates the first The variance of the brightness values ​​of pixels on a fabric image; Indicates the first Information entropy of pixel brightness values ​​in a fabric image; Indicates the first Brightness distribution characteristic values ​​of a fabric image; This represents the mean of the brightness distribution characteristic values ​​of all fabric images; Indicates the complexity of the fabric surface texture;

[0046] variance The larger, the more The more dispersed the distribution of pixel brightness values ​​on a fabric image, the more obvious the alternation of light and dark in the texture; information entropy The larger, the more The more disordered the distribution of pixel brightness values ​​on a fabric image, the greater the texture complexity. The larger the value, the more complex the texture of the fabric surface. At the same time, normalizing the mean of the brightness distribution characteristic value is to limit the value of the target color data to a reasonable range in the subsequent process of correcting the printing template image.

[0047] Step S4: Based on the gloss and texture complexity of the fabric surface, correct the color data of the printing template image to obtain the corrected target color data;

[0048] Correcting the color presentation of prints on fabric involves using mathematical transformations to simulate the actual impact of gloss and texture complexity on human color perception. This ensures that when the printed pattern is reproduced on the fabric, the final printed product maintains a high degree of color visual consistency with the digital image of the printed template provided by the customer, thus resolving the perceptual issues caused by the optical properties of the fabric.

[0049] Specifically, the gloss and texture complexity of the fabric surface are obtained, the brightness data of the printing template image is extracted as the base brightness, and the brightness of the printing template image is compensated and corrected accordingly based on the influence of the fabric surface gloss and texture complexity on color presentation, so that the color effect after printing is more in line with the actual scene observed by the human eye.

[0050] For high-gloss fabrics, such as silk, satin, and synthetic leather, there are highly reflective mirror-like areas. When light shines directly on the corresponding areas, bright highlights are produced, and the brightness of the local area is significantly increased, making the overall fabric print look brighter and more vibrant.

[0051] For low-gloss or matte fabrics, such as cotton canvas, wool flannel, and untreated linen, light undergoes uniform diffuse reflection, resulting in a more even distribution of brightness on the fabric surface and the absence of strong highlights. Due to the multiple reflections and absorptions of light in the micro-texture, the overall reflectivity of the fabric is relatively low. Compared to high-gloss fabrics of the same color, the print brightness is often lower, and the print colors appear more "dull".

[0052] For fabrics with complex textures, such as tweed, corduroy, jacquard fabrics, and coarse knit fabrics with rough surfaces, the uneven textures create a large number of subtle shadows and shading areas. These shadow areas absorb light, which significantly reduces the overall brightness of the fabric print.

[0053] For fabrics with simple textures, such as fine plain cotton, worsted wool, and smooth silk, the surface is flatter, there are fewer shadows, and the light reflects more directly. The overall brightness of the fabric print is higher, and the distribution of print colors is more uniform.

[0054] The specific formula for correcting the brightness of the printing template image based on gloss and texture complexity is as follows:

[0055]

[0056] In the formula, This represents the brightness data of the corrected target color, where L represents the base brightness of the printing template image. Indicates the complexity of the fabric surface texture. Indicates the glossiness of the fabric surface. Indicates the diffuse reflection suppression coefficient. Indicates the specular highlight correction factor;

[0057] Diffuse reflection suppression coefficient Its purpose is to control texture complexity For basic brightness The complementary effect is that the more complex the texture, the stronger the diffuse reflection effect on the fabric surface, and the more obvious the suppression of brightness is during actual printing. Therefore, the complexity of the texture... The larger the size, the more important it is to increase the base brightness. The value is used to achieve the basic brightness. Supplementary corrections, The range of values ​​is generally 100. Specular highlight correction factor Its purpose is to control gloss. The degree of suppression of specular highlights depends on the smoothness of the fabric surface; the stronger the specular reflection effect, the more pronounced the enhancement of brightness during actual printing. Therefore, gloss... The larger the size, the more necessary it is to reduce the base brightness. The value is used to achieve the basic brightness. Suppression correction, The range of values ​​is generally 100. Diffuse reflection suppression coefficient and specular highlight correction factor The specific value can be adjusted according to the material characteristics of the fabric in the actual application scenario.

[0058] In addition, the gloss and texture complexity of the fabric surface are obtained, and the color component data of the printing template image is extracted as the basic color component. The color component data includes the basic color component a from green to red and the basic color component b from blue to yellow. Based on the influence of the gloss and texture complexity of the fabric surface on the color presentation, the saturation of the printing template image is compensated and corrected accordingly so that the color presentation effect after printing is more in line with the actual scene observed by the human eye.

[0059] For high-gloss fabrics, such as silk, satin, and synthetic leather, the color saturation will be significantly reduced in the strongly reflective mirror area, even approaching white or the color of the light source, presenting a light or almost "faded" effect.

[0060] For low-gloss, matte fabrics, such as cotton canvas, wool flannel, and untreated linen, the "true" saturation of colors can be more directly and consistently reflected, without causing local "color loss" due to strong highlights.

[0061] For fabrics with complex textures, such as tweed, corduroy, jacquard fabrics, and coarse knit fabrics with rough surfaces, the visual mixing caused by the subtle color mixing in the highlight and shadow areas can lead to a reduction in the overall perceived saturation. At the same time, the microscopic scattering of the textured surface makes the path of light entering the human eye more complex, which may include more ambient stray light or cause interference between colored lights. Ambient stray light and interference with colored lights can affect the presentation of saturation.

[0062] For fabrics with simple textures, such as fine plain cotton, worsted wool, and smooth silk, the flat surface is the best carrier to showcase the color potential of dyes or pigments. Without complex shadows and color light interference, the color saturation can be fully and purely expressed, making the printed colors appear uniform, clean, and vibrant.

[0063] The specific formula for correcting the saturation of the printing template image based on gloss and texture complexity is as follows:

[0064]

[0065]

[0066] In the formula, This represents the 'a' component data of the corrected target color, where 'a' represents the basic color components from green to red in the printing template image. This represents the b-component data of the corrected target color, where b represents the basic color components from blue to yellow in the printing template image. Indicates the glossiness of the fabric surface. Indicates the complexity of the fabric surface texture. Indicates the specular reflection saturation enhancement coefficient. Indicates the diffuse reflection saturation enhancement coefficient;

[0067] Specular reflection saturation enhancement coefficient Its purpose is to control gloss. For color components and color components The enhancement effect of saturation, due to gloss The higher the gloss level, the stronger the specular reflection. When the human eye observes the actual printed image under high light, the more intense the light perceived, causing the color observed by the human eye to be closer to white or the color of the light source. This reduces the perceived color saturation of the printed image, thus affecting the gloss level. The larger the value, the more necessary it is to increase the value of the basic color components in order to supplement and correct the basic color components a and b. The range of values ​​is generally 100. Diffuse reflection saturation enhancement coefficient Its purpose is to control texture complexity For color components and color components The enhancement of saturation is negatively impacted by complex textures, which can lead to visual mixing, ambient light interference, and color light glare, thus affecting the perceived saturation of printed colors. Therefore, texture complexity... The larger the value, the more necessary it is to increase the value of the basic color components in order to supplement and correct the basic color components a and b. The range of values ​​is generally 100. Specular reflection saturation enhancement coefficient and diffuse reflection saturation enhancement coefficient The specific value can be adjusted according to the material characteristics of the fabric in the actual application scenario and the application environment of the printed fabric.

[0068] Step S5: Calculate the theoretical spectral reflectance corresponding to different mixing ratios of inks to obtain the predicted color data of the mixed inks;

[0069] Specifically, according to Theoretical calculations were performed to obtain the theoretical spectral reflectance of different ink mixture ratios on an ideal fabric, and the results were then used to find the appropriate standard light source model. Standard documentation, the standard light source is usually selected as D65, to obtain the standard observer function. , , Spectral power distribution function of light source Reference white points value , , The theoretical spectral reflectance is converted into a standard observer function and a light source spectral power distribution function. Tristimulus values, then referenced by white points value pairs The tristimulus values ​​were normalized to obtain the predicted color values ​​of the mixed inks with corresponding ratios on the ideal fabric. The predicted color data of mixed inks with different ratios on the ideal fabric were then calculated sequentially. ;

[0070] according to The theoretical spectral reflectance of the mixed ink on an ideal fabric is obtained theoretically, and the specific calculation formula is as follows:

[0071]

[0072]

[0073]

[0074] In the formula, , , , It is the normalized concentration of the ink mixing parameters, which satisfies and ; , , , Indicates the absorption coefficient of a single-color ink. , , , The scattering coefficient of a single-color ink is represented by printing a single-color patch and measuring its spectral reflectance, then combined with... The equation is obtained by reverse calculation; , Indicates the absorption and scattering coefficients of the mixed ink; , The absorption and scattering coefficients of an ideal fabric can be obtained by measuring the monochromatic reflectance of the template fabric substrate using a spectrophotometer. This represents the spectral reflectance of an ideal fabric. This represents the theoretical spectral reflectance of the mixed ink on an ideal fabric.

[0075] Step S6: Based on the target color data of the corrected printing template image, match the predicted color with the smallest color difference between it and the target color to obtain the corresponding optimal ink ratio parameters;

[0076] Specifically, adopt The function constructs a color difference minimization objective function or a global search algorithm to find the predicted color with the smallest color difference from the target color as the final recommended color. Based on the mapping relationship between ink ratio and predicted color, the ink ratio parameter corresponding to the predicted color with the smallest color difference is obtained and used as the optimal ink ratio parameter. The optimal ink ratio parameter is used as the execution printing parameter of the inkjet printer and output to the inkjet printer for printing.

[0077] The specific formula for calculating the color difference between the predicted color and the target color is as follows:

[0078]

[0079] In the formula, The value representing the color difference. Data for the target color; Data for predicting colors.

[0080] This invention also discloses an intelligent interactive system based on inkjet printing color recommendation, the structure of which is as follows: Figure 2 As shown, the intelligent interactive method for implementing the above-mentioned inkjet printing color recommendation includes: an inkjet printing device, a processor, a memory, a communication interface, a standard light source, and an image acquisition device. The processor stores computer program instructions for implementing the above-mentioned intelligent interactive method. The image acquisition device can be adjusted in multiple directions and angles. The standard light source can simulate the illumination environment of natural light. The communication interface is communicatively connected to the image acquisition device and the inkjet printing device.

[0081] The embodiments included in this invention are merely descriptions of preferred embodiments of the invention and are not limited to the precise structures described above and shown in the accompanying drawings. Various modifications and changes can be made without departing from the scope of protection. Any variations and improvements made by those skilled in the art to the technical solutions of this invention without departing from the design concept of this invention should fall within the scope of protection of this invention.

Claims

1. An intelligent interaction method based on inkjet printing color recommendation, characterized in that: The fabric images and the printing template images photographed from multiple angles are obtained and converted into the format of Lab color space, to obtain the brightness value data of each image; based on the changes of the brightness values between the fabric images, the glossiness of the fabric surface is quantified; based on the distribution of the brightness values between the pixel points on each fabric image, the texture complexity of the fabric surface is quantified; based on the glossiness and the texture complexity of the fabric surface, the color data of the printing template image is corrected, to obtain the target color data after correction; The theoretical spectral reflectivity corresponding to the mixed ink with different proportions is calculated, to obtain the predicted color data of the mixed ink; based on the target color data of the printing template image after correction, the predicted color with the smallest color difference from the target color is matched, to obtain the optimal ink proportion parameter corresponding to the predicted color. The brightness component of the printing template image is adjusted by making up for the diffuse reflection attenuation of the brightness visual effect caused by the texture complexity of the fabric and by inhibiting the highlight effect caused by the glossiness of the fabric, so that the brightness correction of the printing template image is realized; the color component of the printing template image is adjusted by reducing the negative influence of the visual mixing and the mutual interference of color lights caused by the texture complexity of the fabric on the saturation presentation effect and by inhibiting the influence of the reflected light caused by the glossiness of the fabric on the saturation visual effect, so that the saturation correction of the printing template image is realized.

2. The intelligent interaction method for inkjet printing color recommendation according to claim 1, characterized in that, The preparation of the fabric sample to be printed, the arrangement of the standard light source simulating natural light and the vertical angle between the light source and the fabric, the multi-directional and multi-angle shooting of the fabric by the camera lens around the fabric under the irradiation of the standard light source, the determination of the shooting angle by the included angle between the camera lens and the normal line of the fabric surface, the collection of the images of the fabric at different angles in the unprinted state, the conversion of the multi-angle fabric images and the printing template image into the format of Lab color space, and the obtaining of the brightness value data of all the pixel points of each image are included.

3. The intelligent interaction method for inkjet printing color recommendation according to claim 1, characterized in that, The glossiness of the fabric surface is quantified based on the change of the brightness value between each fabric image, including: based on the brightness value data of each fabric image, obtaining the number of different angles used when shooting the fabric image, calculating the average brightness value of all fabric images under each shooting angle, then calculating the mean value of the brightness value of all fabric images under all shooting angles, and the variance of the average brightness value under all shooting angles, using logarithmic change to compress the value of the variance and enhance the non-linear characteristics of the brightness value between each fabric image, obtaining the glossiness characteristic value of the fabric surface, using The function is used for normalization processing of the glossiness characteristic value, and the normalized value of the glossiness characteristic value is used as the glossiness of the fabric surface.

4. The intelligent interaction method for inkjet printing color recommendation according to claim 1, characterized in that, Based on the brightness value data of each fabric image, the total number of fabric images, the number of pixel points of each fabric image and the brightness value of each pixel point are obtained, the average brightness value of each fabric image is calculated, the value range of the brightness value, i.e. 1 to 100, is divided into multiple equal-length intervals, the pixel points are divided into the intervals based on the size of the brightness value, the number of pixel points in each interval on each fabric image is counted, and the proportion of the number of pixel points in each interval on the corresponding fabric image is calculated.

5. The intelligent interaction method for inkjet printing color recommendation according to claim 4, characterized in that, The quantification of the texture complexity of the fabric surface based on the distribution of the brightness values between the pixel points on each fabric image further comprises: calculating the variance and information entropy of the brightness values of the pixel points on each fabric image, adding the variance and information entropy corresponding to each fabric image to obtain a brightness distribution characteristic value of the corresponding fabric image, calculating the mean value of the brightness distribution characteristic values of all fabric images, and normalizing the mean value of the brightness distribution characteristic values by using a function to obtain a normalized value as the texture complexity of the fabric surface. The function is used to normalize the mean value of the brightness distribution characteristic values, and the normalized value is used as the texture complexity of the fabric surface.

6. The intelligent interaction method for inkjet printing color recommendation according to claim 1, characterized in that, The method comprises the following steps: obtaining the glossiness and texture complexity of the fabric surface, extracting the brightness data of the printing template image as the basic brightness, setting the diffuse reflection suppression coefficient and the specular highlight correction coefficient according to the material characteristics of the fabric in the actual application scene, taking the product of the texture complexity and the diffuse reflection suppression coefficient as the brightness supplementary coefficient, taking the product of the glossiness and the specular highlight correction coefficient as the brightness suppression coefficient, taking the difference between the brightness supplementary coefficient and the brightness suppression coefficient as the brightness correction coefficient, taking the product of the brightness correction coefficient and the basic brightness as the brightness correction value, and taking the sum of the brightness correction value and the basic brightness as the brightness value in the corrected target color data.

7. The intelligent interaction method for inkjet printing color recommendation according to claim 6, characterized in that, The method further comprises the following steps: obtaining the glossiness and texture complexity of the fabric surface, extracting the color component data of the printing template image as the basic color component, setting the specular reflection saturation enhancement coefficient and the diffuse reflection saturation enhancement coefficient according to the material characteristics of the fabric and the application environment of the printed fabric in the actual application scene, taking the product of the glossiness and the specular reflection saturation enhancement coefficient as the saturation enhancement coefficient, taking the product of the texture complexity and the diffuse reflection saturation enhancement coefficient as the saturation supplementary coefficient, taking the sum of the saturation enhancement coefficient and the saturation supplementary coefficient as the saturation correction coefficient, taking the product of the saturation correction coefficient and the basic color component as the saturation correction value, and taking the sum of the saturation correction value and the basic color component as the color component value in the corrected target color data.

8. The intelligent interaction method for inkjet printing color recommendation according to any one of claims 1 to 7, characterized in that, The calculation of the theoretical spectral reflectance corresponding to different ratios of mixed inks to obtain the predicted color data of the mixed inks includes: based on Theoretical calculations were performed to obtain the theoretical spectral reflectance of different ink mixture ratios on an ideal fabric, and the results were then used to search for the selected standard light source model. Standard documentation provides the standard observer function, the light source spectral power distribution function, and the reference white point. The theoretical spectral reflectance is converted to a value based on the standard observer function and the spectral power distribution function of the light source. Tristimulus values, then referenced by white points value pairs The tristimulus values ​​are normalized to obtain the predicted color values ​​of the mixed inks with the corresponding ratios on the ideal fabric. The predicted color data of the mixed inks with different ratios on the ideal fabric are then calculated sequentially.

9. The intelligent interaction method for inkjet printing color recommendation according to claim 8, characterized in that, The target color data based on the corrected stamp template image matches a predicted color with the smallest color difference from the target color, to obtain corresponding optimal ink ratio parameters, including: The function constructs a color difference minimization target function or a global search algorithm, finds a predicted color with the smallest color difference from the target color as the final recommended color, obtains the ink ratio parameters corresponding to the predicted color when the color difference is the smallest based on the mapping relationship between the ink ratio and the predicted color, and takes the ink ratio parameters as the optimal ink ratio parameters. The optimal ink ratio parameters are taken as the execution printing parameters of the inkjet printing device, and output to the inkjet printing device for printing.

10. An intelligent interactive system for inkjet printing color recommendation, characterized in that: The system comprises an inkjet printing device, a processor, a memory, a communication interface, a standard light source and an image acquisition device. The processor stores computer program instructions for implementing the intelligent interaction method of any one of claims 1 to 9. The image acquisition device can be adjusted in multiple directions and multiple angles. The standard light source can simulate the illumination environment of natural light. The communication interface is in communication connection with the image acquisition device and the inkjet printing device.

Images