A soft proofing method for printed fabrics based on image processing

Through image processing technology, the Mean-shift algorithm and Pantone color library are used to solve the problem of cumbersome color testing and proofing of printed fabrics, and efficient soft proofing of printed fabrics is achieved, providing clear color change results, saving resources and shortening delivery cycle.

CN113192044BActive Publication Date: 2024-11-05JIANGNAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110522191.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-05-13
Publication Date
2024-11-05
Estimated Expiration
2041-05-13

AI Technical Summary

Technical Problem

When producing series of printed fabrics, the cumbersome color test and proofing process is time-consuming and cost-effective, and the existing computer image segmentation technology cannot effectively process the texture characteristics of fabric images, resulting in color defects in color change, especially for fabrics with complex texture colors. The rate is affected.

Method used

The soft proofing method of printed fabrics based on image processing is adopted, and the smooth texture features of the image are preprocessed, and the image segmentation is performed using the Mean-shift clustering algorithm. Combined with the Pantone color library, the color matching scheme is selected, and the mapping relationship is established to realize the color change of the image.

Benefits of technology

It realizes efficient computer soft proofing of printed fabrics, with good edge details segmentation effect, saves manpower and raw materials, shortens the delivery cycle, and provides series color change results with clear and natural structural texture.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113192044B_ABST
    Figure CN113192044B_ABST
Patent Text Reader

Abstract

The present invention belongs to the technical field of printed fabric proofing, and specifically relates to a printed fabric soft proofing method based on image processing. The method provided by the present invention can realize computer soft proofing of printed fabrics through image processing technology, and the adopted Mean‑shift algorithm can adapt to different types of fabric images, and the segmentation effect of edge detail parts is distinguishable. By adopting the method for soft proofing of printed fabrics, a series of color change results with clear and natural structural textures can be obtained, and a color test reference can be provided. Compared with the traditional machine proofing method, it saves manpower and raw materials and shortens the delivery cycle.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of printed fabric proofing, and in particular relates to a printed fabric soft proofing method based on image processing. Background Art

[0002] With the rapid development of the times, consumers' demands for products tend to be diversified. In addition to practical performance, they are more pursuing novel and popular appearance. In the new era of rapid product replacement, printed fabrics are deeply loved by consumers for their rich and varied colors and patterns. Before enterprises mass-produce printed fabrics, they need professional designers to design patterns and determine color schemes, and production personnel to verify the design schemes by making plates and proofing according to the design schemes. Printing and dyeing proofing has always been a time-consuming, high-cost and unavoidable process. In order to achieve satisfactory results, proofing is often required repeatedly.

[0003] With the rapid development of computer image processing technology, its application in the textile field is becoming more and more extensive. It has become a trend to use computer image processing technology to perform machine soft proofing on printed fabrics. Although computer image segmentation technology, such as K-means clustering method, region growing method, etc., can be used to achieve image segmentation, and image color change can be achieved through methods such as color migration. However, fabric images have unique texture characteristics, which will affect the image segmentation results and cause color defects in the color change results. In addition, many segmentation algorithms cannot perform detailed segmentation on the details of fabric images. For fabrics with complex textures and colors, the accuracy of color change results will be greatly affected. Summary of the invention

[0004] The invention aims to solve the cumbersome color trial proofing process of printed fabrics when producing series products, and proposes a soft proofing method for printed fabrics based on image processing.

[0005] The technical solution of the present invention is as follows:

[0006] A soft proofing method for printed fabrics based on image processing, the steps are as follows:

[0007] Step 1: The user uploads the printed fabric image that he wants to change the color to the soft proofing system;

[0008] Step 2: Preprocess the input image and perform color space conversion to smooth the texture features of the fabric;

[0009] Furthermore, the image preprocessing method adopts a relative total variation model, and suppresses the texture information of the fabric image by adjusting the smoothing parameter λ and the window size parameter σ of the model.

[0010] Preferably, the smoothing parameter λ is set to 0.006, and the window size parameter σ is set to 4.

[0011] Furthermore, the color space conversion is to convert the preprocessed image from the RGB color space to the CIELAB color space for image segmentation.

[0012] Step 3: Use clustering algorithm to segment the preprocessed image and extract the color separation of printed fabrics;

[0013] Preferably, the cluster segmentation algorithm adopts the Mean-shift algorithm, and the steps include making a drift window with a bandwidth h, calculating the offset required for the center point to move to the surrounding points in the window, obtaining the average offset, drifting the original center point in the direction of the average offset to obtain a new center point, repeating the above steps until the algorithm converges, and obtaining the cluster center and the corresponding segmentation area.

[0014] Preferably, the bandwidth parameter h is set to 0.11.

[0015] Furthermore, the printed fabric color separation extraction method is to extract the color separation information of the input image according to the values ​​of the cluster centers of the three channels L, a, and b, and obtain the cluster segmentation map of the input image according to the cluster segmentation result.

[0016] Furthermore, based on the segmentation result, the area where the color is to be changed is found.

[0017] Step 4: Select a color scheme from the Pantone color library;

[0018] Step 5: Pair the selected new color with the color separation obtained by segmentation extraction, establish a mapping relationship, map the new color to the segmented area corresponding to the input image, and output the final color change result;

[0019] Furthermore, the pairing method is that the newly selected color scheme and the color separation are in a one-to-one correspondence.

[0020] Furthermore, the method for establishing the mapping relationship is to map the color matching value to the cluster segmentation area corresponding to the color separation according to the corresponding relationship between the color matching and the color separation in the CIELAB color space. Then, the mapping result is converted from the CIELAB color space to the RGB color space, and the final color-changed image is output;

[0021] Step 6: Save the color change soft proofing result image and save the Pantone color library value of the corresponding color scheme to provide reference for subsequent production.

[0022] Furthermore, the color matching scheme of the present invention is associated with the Pantone color library. After the soft proofing result meets the customer's requirements, the color parameters and the original process sheet can be directly called to guide production.

[0023] Beneficial effects of the present invention:

[0024] The method provided by the present invention can realize computer soft proofing of printed fabrics through image processing technology. The adopted Mean-shift algorithm can adapt to different types of fabric images, and the segmentation effect of edge details is distinguishable. The method is used for soft proofing of printed fabrics to obtain a series of color change results with clear and natural structure textures, and can provide a color test reference. Compared with the traditional machine proofing method, it saves manpower and raw materials and shortens the delivery cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention is a flowchart of a soft proofing method for printed fabrics based on image processing according to a preferred embodiment of the present invention.

[0026] Figure 2 : are the original image and the smoothed image of the fabric in the embodiment.

[0027] Figure 3 This is a color separation extraction diagram after image segmentation in the embodiment.

[0028] Figure 4 This is a color separation and color change diagram of an embodiment image.

[0029] Figure 5 Series product images simulated for soft proofing. DETAILED DESCRIPTION

[0030] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0031] Figure 1 The present invention is a flowchart of a method for soft proofing of printed fabrics based on image processing according to an embodiment of the present invention.

[0032] The method of this embodiment includes the following steps:

[0033] Step 1: The user uploads the printed fabric image that he wants to change the color to the soft proofing system;

[0034] In this step, the present invention collects different printed fabrics and uploads the fabric images to the soft proofing system in sequence to perform a color change experiment.

[0035] Step 2: Preprocess the input image to smooth the texture features of the fabric;

[0036] In this step, the image preprocessing method uses the relative total variation model, and the texture information of the fabric image is suppressed by adjusting the smoothing parameter λ and the window size parameter σ of the model. The specific objective function is:

[0037]

[0038] Where i, j represents the position of the pixel in the image, S(i, j) represents the input image, T(i, j) represents the structure extraction map, argmin represents the set of points (i, j) when the latter formula reaches the minimum value, (T(i, j)-S(i, j)) 2 Ensure that the difference between input and output is minimal. Select the optimal parameters and use the relative total variation model to obtain the following Figure 2 The printed fabric image after texture smoothing is shown. The smoothed image is then converted from the RGB color space to the CIELAB color space, which has uniform color and is independent of the sampling environment and equipment, to facilitate subsequent image segmentation.

[0039] Step 3: Use clustering algorithm to segment the preprocessed image and extract the color separation of printed fabrics;

[0040] In this step, the cluster segmentation algorithm used is the Mean-shift algorithm. The algorithm iteration steps are as follows: 1) Make a drift window with bandwidth h, calculate the center point x0 and move it to the surrounding points x in the window i The sum of the required offsets is calculated and the average is obtained to obtain the average offset M(x0); 2) Point x0 drifts in the direction of the average offset to obtain a new center point x0'; 3) Repeat the above steps until the algorithm converges. The offset mean M(x0) is:

[0041]

[0042] The preprocessed embodiment image is segmented by the above method, and the color separation of the image is extracted according to the cluster center values ​​of the three channels L, a, and b. According to the cluster segmentation result, a cluster segmentation map of the input image is obtained. The color separation and segmentation area map obtained by the embodiment segmentation is converted to RGB color space for display, as shown in reference Figure 3 shown.

[0043] Step 4: Select a color scheme from the Pantone color library;

[0044] Select the color you want to replace based on the segmentation result and determine the new color scheme. The newly selected color scheme has a one-to-one correspondence with the color separation, such as Figure 4 shown.

[0045] Step 5: Pair the selected new color with the color separation obtained by segmentation extraction, establish a mapping relationship, map the new color to the segmented area corresponding to the input image, and output the final color change result;

[0046] In this step, the new color matching is mapped to the segmented area corresponding to the color separation of the original image in the CIELAB color space. Assume that the fabric image with a size of M×N is I M×N×3 , fabric clustering segmentation to obtain k categories, ΔCk Indicates the color difference between the kth color separation and the corresponding new color matching, I Ck Locate the position of the k-th color change; I C is the corresponding three-channel value after color change. r is the number of colors to be changed. When r is 1, single-color color change is performed; when r>=1, multi-color color change is performed. The color separation and color change formula is as follows:

[0047]

[0048]

[0049] Then convert the mapped color change result from CIELAB color space to RGB color space, output the final color change image, and refer to the color change example Figure 4 , other soft proofing implementation examples are available at Figure 5 .

[0050] Step 6: Save the color change soft proofing result image and save the Pantone color library value of the corresponding color scheme to provide reference for subsequent production.

[0051] The color matching scheme of the present invention is associated with the Pantone color library. After the soft proofing result meets the customer's requirements, the color parameters and the original process sheet can be directly called to guide production.

[0052] Those skilled in the art should understand that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A soft proofing method for printed fabrics based on image processing, characterized in that: Here are the steps: Step 1: The user uploads the image of the printed fabric that he wants to change the color of; Step 2: Preprocess the input image and perform color space conversion to smooth the texture features of the fabric; The image preprocessing method adopts a relative total variation model, and suppresses the texture information of the fabric image by adjusting the smoothing parameter λ and the window size parameter σ of the model; The color space conversion is to convert the preprocessed image from the RGB color space to the CIELAB color space for image segmentation; Step 3: Use clustering algorithm to segment the preprocessed image and extract the color separation of printed fabrics; The clustering algorithm adopts the Mean-shift algorithm, which includes the following steps: making a drift window with a bandwidth parameter h, calculating the offset required for the center point to move to the surrounding points in the window, obtaining the average offset, drifting the original center point in the direction of the average offset to obtain a new center point, repeating the Mean-shift algorithm until the algorithm converges, and obtaining the cluster center and the corresponding segmentation area; Step 4: Select a color scheme from the Pantone color library; Step 5: Pair the selected new color with the color separation obtained by segmentation extraction, establish a mapping relationship, map the new color to the segmented area corresponding to the input image, and output the final color change result; The pairing method is that the newly selected color scheme and the color separation are in a one-to-one correspondence; Step 6: Save the color change soft proofing result image and save the Pantone color library value of the corresponding color scheme to provide reference for subsequent production.

2. A printing fabric soft proofing method based on image processing as claimed in claim 1, characterized in that: In step 2, the smoothing parameter λ is set to 0.006, and the window size parameter σ is set to 4.

3. A printing fabric soft proofing method based on image processing as claimed in claim 1 or 2, characterized in that: In step 3, the bandwidth parameter h is set to 0.

11.

4. A printing fabric soft proofing method based on image processing as claimed in claim 1 or 2, characterized in that: In step 3, the method for extracting color separation of printed fabrics is to extract the color separation information of the input image according to the values ​​of the cluster centers of the three channels L, a, and b, and obtain the cluster segmentation map of the input image according to the cluster segmentation result.

5. A printing fabric soft proofing method based on image processing as claimed in claim 3, characterized in that: In step 3, the method for extracting color separation of printed fabrics is to extract the color separation information of the input image according to the values ​​of the cluster centers of the three channels L, a, and b, and obtain the cluster segmentation map of the input image according to the cluster segmentation result.

6. A printing fabric soft proofing method based on image processing as claimed in claim 1, 2 or 5, characterized in that: In step 5, the method for establishing the mapping relationship is to map the color matching value to the cluster segmentation area corresponding to the color separation according to the correspondence between the color matching and the color separation in the CIELAB color space; then convert the mapping result from the CIELAB color space to the RGB color space, and output the final color-changed image.

7. A printing fabric soft proofing method based on image processing as claimed in claim 3, characterized in that: In step 5, the method for establishing the mapping relationship is to map the color matching value to the cluster segmentation area corresponding to the color separation according to the correspondence between the color matching and the color separation in the CIELAB color space; then convert the mapping result from the CIELAB color space to the RGB color space, and output the final color-changed image.

8. A printing fabric soft proofing method based on image processing as claimed in claim 4, characterized in that: In step 5, the method for establishing the mapping relationship is to map the color matching value to the cluster segmentation area corresponding to the color separation according to the correspondence between the color matching and the color separation in the CIELAB color space; then convert the mapping result from the CIELAB color space to the RGB color space, and output the final color-changed image.