A method for direct-to-garment printing with organic plant dyes
By using image processing and data acquisition modules, combined with neural network models and deep learning algorithms, the residence time of the printhead at the printing point is adjusted in real time, solving the problem of poor printing quality in existing technologies and achieving high-quality and stable printing effects.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAOXING SHIZHIRAN FASHION TECHNOLOGY CO LTD
- Filing Date
- 2024-05-08
- Publication Date
- 2026-05-26
AI Technical Summary
Existing direct-to-garment printing technology struggles to adjust the printhead's dwell time at the printing point in a timely manner based on the material and thickness of the object being printed, as well as the temperature during printing, resulting in poor printing quality.
Through image processing and data acquisition modules, the residence time of the printhead at the printing point is adjusted in real time. Combined with neural network models and deep learning algorithms, the residence time of the printhead at the printing point and the dye penetration depth are corrected in real time according to fabric classification, thickness and color blocks.
It improves printing quality and accuracy, ensures the consistency and stability of printing effects, and avoids printing quality problems caused by excessively long or short dwell times.
Smart Images
Figure CN118560183B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of digital inkjet printing and relates to direct inkjet printing technology using organic plant dyes, specifically a direct inkjet printing method using organic plant dyes. Background Technology
[0002] Traditional textile printing processes often use chemically synthesized dyes, whose production and use cause environmental problems. Organic plant dyes, on the other hand, are naturally derived, free of harmful substances, environmentally friendly, and skin-friendly. Printing techniques offer advantages such as personalization, sample systems, and high production efficiency. Combining organic plant dyes with printing technology can achieve more environmentally friendly, innovative, and higher-quality printing results. Direct-to-garment printing with organic plant dyes involves spraying dyes directly onto textiles to create the printed pattern. This method aims to achieve environmentally friendly, sustainable, and high-quality textile printing technology.
[0003] In existing direct-to-garment printing technologies, the dwell time of the printhead at the printing point is mostly preset. It is difficult to adjust the dwell time of the printhead at the printing point in a timely manner according to the material and thickness of the printing object and the temperature during printing. After the printhead has been used for a long time, the temperature of the printhead will rise, which will lead to the dye temperature. Different fabrics have different affinity for dyes. A fixed dye printing time will affect the adaptability to different printing objects and make it difficult to guarantee the printing quality. Summary of the Invention
[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a direct-to-garment printing method for organic plant dyes, which solves the technical problem of difficulty in timely adjusting the residence time of the printhead at the printing point according to the material and thickness of the printing object and the temperature during printing and dyeing.
[0005] To address the above problems, a first aspect of the present invention provides a method for direct-to-garment printing with organic plant dyes, comprising the following steps:
[0006] Import the pattern file to be printed into the image processing module. Based on the printing area of the object to be printed, scale and rotate the pattern to be printed, and position the printing area of the object to be printed, positioning the printed pattern to the printing area of the object to be printed.
[0007] Input the fabric type of the printing area of the object to be printed, assign the fabric type of the printing area to the direct-to-garment printing fabric category, and detect the fabric thickness.
[0008] Based on the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, the time the printhead stays at each printing point in the printing area is preset.
[0009] The printed pattern is divided into blocks of different colors, and these blocks are further classified into dark and light blocks. The printing is then performed on the different blocks.
[0010] Based on the color of the direct-to-garment printing blocks, the classification of different direct-to-garment printing fabrics, and the thickness of the fabrics, the preset values of the dwell time of the print head at each printing point in different printing and dyeing areas are adjusted in real time.
[0011] The dye immersion depth of the finished direct-to-garment printed fabric and the preset dye immersion depth data are sampled and evaluated. Based on the evaluation results, the calculation formula for the preset time of the print head staying at each printing point in the printing area is revised.
[0012] As a further aspect of the present invention: the direct-to-garment printing fabrics are classified as follows: natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics.
[0013] As a further aspect of the present invention: based on the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, the dwell time of the printhead at each printing point in the printing area is preset, including the following steps:
[0014] Based on the printing area of the object to be printed, direct-print fabrics are classified into natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics.
[0015] The time the printhead spends at each printing point in the printing area is preset using the following formula:
[0016]
[0017] Where T is the time the nozzle stays at each printing point in the printing area, h is the fabric thickness in millimeters, and k is the fabric surface temperature during direct-to-garment printing.
[0018] When the fabric is made of natural animal fiber, α is 0.5 and β is 12; when the fabric is made of natural plant fiber, α is 0.55 and β is 15; when the fabric is made of blended fabric, α is 0.6 and β is 18; when the fabric is made of synthetic fiber, α is 0.6 and β is 20.
[0019] As a further aspect of the present invention: the printed pattern is divided into blocks of different colors, and the blocks of different colors are classified into dark blocks and light blocks, and the different blocks are printed, including the following steps:
[0020] The colors of the pattern to be printed are extracted, and the organic plant dyes are adjusted according to the colors extracted from the pattern.
[0021] The Canny edge detection algorithm is used to detect the color change edges on the pattern to be printed at the boundary positions of different color blocks. The color change edge lines are used as the dividing lines of the color blocks to divide the pattern to be printed into different blocks.
[0022] Colors are picked up from different areas of the pattern to be printed, and the colors picked up from different areas are matched with the colors extracted from the pattern to be printed.
[0023] In the CMYK color model, the K channel values of different blocks that match the color are detected. Blocks with K channel values less than a preset threshold are marked as light blocks, and blocks with K channel values greater than or equal to the preset threshold are marked as dark blocks.
[0024] Based on the matched colors, print on the light and dark areas respectively.
[0025] As a further aspect of the present invention: based on the color of the direct-to-garment printing blocks, the classification of different direct-to-garment printing fabrics, and the thickness of the fabric, the preset value of the dwell time of the printhead at each printing point in different printing areas of the fabric classification is corrected in real time, including the following steps:
[0026] The data acquisition module collects infrared thermal images of the printing area of the object to be printed on fabrics of different thicknesses in different direct-to-garment printing fabric categories during the printing process.
[0027] Organic plant dyes were fluorescently labeled, and the residence time of the dyeing dots and the fluorescence imaging data of the fabrics were collected in different color block categories of fabrics with different thicknesses and different direct-to-garment printing fabrics.
[0028] Train a neural network model based on the data collected by the data acquisition module;
[0029] The system collects infrared thermal imaging images of the printing and dyeing areas in real time, as well as data on direct-to-garment printing fabric classification, fabric thickness, and preset dye immersion depth. Through a trained neural network model, it corrects in real time the preset value of the time the nozzle stays at each printing point in different printing and dyeing areas for each fabric classification.
[0030] As a further aspect of the present invention: Organic plant dyes are fluorescently labeled, and data on the residence time of the dyed dots and the fluorescence imaging of the fabric are collected from different color blocks of different thicknesses of fabrics in different direct-to-garment printing fabric categories. This includes the following steps:
[0031] Fluorescent labeling of plant proteins for organic plant dyes;
[0032] Organic plant dyes with fluorescent markings in dark areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics, while organic plant dyes with fluorescent markings in light areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics.
[0033] The fluorescently labeled organic plant dye is excited by an excitation light source to produce fluorescence. A fluorescence microscope is used to image the fabric, capture the fluorescence signal emitted by the fabric, and generate fluorescence imaging data of the fabric surface.
[0034] The dye penetration depth of a portion of the fabric was detected. Using a deep learning algorithm, the organic plant dye penetration depth data of all data was obtained based on the fluorescence imaging data of the fabric surface.
[0035] As a further aspect of the present invention: the dye penetration depth of a portion of the fabric is detected, and organic plant dye penetration depth data for all data is obtained using a deep learning algorithm based on fluorescence imaging data of the fabric surface, including the following steps:
[0036] The depth of dye penetration in some fabrics was detected, and fluorescence imaging data of the fluorescent dyes were obtained. The fluorescence imaging data of dark and light color areas were divided into two groups.
[0037] Preprocessing of fluorescence imaging data includes image resizing, grayscale conversion, and noise removal;
[0038] Convolutional neural network models were trained separately to learn the relationship between dye penetration depth and fluorescence imaging images when the dark and light areas of direct inkjet printing were dyed.
[0039] The fluorescence imaging data of the fabric surface is input into the convolutional neural network model of the corresponding dark and light color blocks of direct inkjet printing, and the dye penetration depth data corresponding to the fluorescence imaging data of the fabric surface is output.
[0040] As a further aspect of the present invention: training a neural network model based on data collected by the data acquisition module includes the following steps:
[0041] Acquire infrared thermal imaging images of the fabric thickness, the time the nozzle stays at the dyeing point, the dyeing area during dyeing, and data on the dye penetration depth.
[0042] The fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data are used as inputs, and the corresponding nozzle dwell time at the dyeing point is used as the output. CNN neural network models are trained for natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics respectively. These models are used to obtain suggested values for the nozzle dwell time at each dyeing point based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data.
[0043] As a further aspect of the present invention: A preset value for the dwell time of the nozzle at each dyeing point in different fabric categories within different dyeing areas is corrected in real time using a trained neural network model, including the following steps:
[0044] Based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye immersion depth data, the trained CNN neural network model obtains the suggested value of the nozzle dwell time at each dyeing point.
[0045] Calculate the difference between the preset value of the time the printhead stays at each dyeing point in different dyeing areas and the suggested value of the time the printhead stays at each dyeing point. If the difference is less than the preset difference threshold, the time the printhead stays at each dyeing point will not be modified. Otherwise, the time the printhead stays at each dyeing point will be modified to the suggested value of the time the printhead stays at each dyeing point.
[0046] As a further aspect of the present invention: The dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data are randomly sampled and evaluated. Based on the evaluation results, the calculation formula for the preset dwell time of the printhead at each printing point in the printing area is revised, including the following steps:
[0047] The following formula is used to conduct random sampling and evaluation of the dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data:
[0048]
[0049] Wherein, G is the sampling evaluation value, L1 is the preset dye immersion depth data, and L2 is the dye immersion depth of the sampled direct-to-garment printed fabric.
[0050] If the sampling evaluation value G is greater than 0.3, the coefficient α in the calculation formula for the preset time of the nozzle staying at each printing point in the printing and dyeing area will be reduced by 0.01.
[0051] If the sampling evaluation value G is less than -0.3, the coefficient α in the calculation formula for the preset time for the nozzle to stay at each printing point in the printing and dyeing area will be increased by 0.01.
[0052] Otherwise, the coefficients in the calculation formula for the preset time the nozzle stays at each printing point in the printing area are not adjusted.
[0053] Compared with the prior art, the beneficial effects of the present invention are:
[0054] This invention pre-sets the dwell time of the printhead at each printing point in the printing area based on the classification of direct-to-garment printing fabrics in the printing area. It then adjusts the pre-set dwell time in real-time according to the color of the printing area, the different types of direct-to-garment printing fabrics, and the fabric thickness. By adjusting the dwell time at each printing point in real-time based on the thickness of different fabrics and the different characteristics of the printing area, the quality and accuracy of the printing are improved. Real-time correction of the dwell time during the printing process avoids printing quality problems caused by excessively long or short dwell times, ensuring the consistency and stability of the printing effect.
[0055] This invention categorizes direct-to-garment printing fabrics according to the printing area to be printed, and presets the dwell time of the printhead at each printing point in the printing area. It then conducts random checks and evaluations of the dye penetration depth of the completed direct-to-garment printing fabric against the preset dye penetration depth data. Based on the evaluation results, the calculation formula for the preset dwell time of the printhead at each printing point in the printing area is revised. By randomly checking and evaluating the dye penetration depth of the completed direct-to-garment printing fabric against the preset dye penetration depth data, it is easier to determine whether the dye penetration depth meets the preset requirements, thus improving the quality of the printing effect. Revising the calculation formula based on the evaluation results helps improve the accuracy of the preset dwell time at the printing points, avoiding excessive adjustments to the printing time at the printing points later. Attached Figure Description
[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0057] Figure 1 This is a schematic diagram of the method flow of the present invention;
[0058] Figure 2 This is a schematic diagram illustrating the process of real-time correction of the preset value of the time the nozzle stays at the printing point according to the present invention. Detailed Implementation
[0059] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0060] Please see Figures 1-2 The first aspect of this invention provides a method for direct-to-garment printing with organic plant dyes, comprising the following steps:
[0061] Import the pattern file to be printed into the image processing module. Based on the printing area of the object to be printed, scale and rotate the pattern to be printed, and position the printing area of the object to be printed, positioning the printed pattern to the printing area of the object to be printed; supports common image formats, such as JPEG, PNG, etc.
[0062] Input the fabric type of the printing area of the object to be printed, assign the fabric type of the printing area to the direct-to-garment printing fabric category, and detect the fabric thickness.
[0063] Based on the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, the time the printhead stays at each printing point in the printing area is preset.
[0064] The printed pattern is divided into blocks of different colors, and these blocks are further classified into dark and light blocks. The printing is then performed on the different blocks.
[0065] Based on the color of the direct-to-garment printing blocks, the classification of different direct-to-garment printing fabrics, and the thickness of the fabrics, the preset values of the dwell time of the print head at each printing point in different printing and dyeing areas are adjusted in real time.
[0066] The dye immersion depth of the finished direct-to-garment printed fabric and the preset dye immersion depth data are sampled and evaluated. Based on the evaluation results, the calculation formula for the preset time of the print head staying at each printing point in the printing area is revised.
[0067] Specifically, in this embodiment, the dwell time of the printhead at each printing point in the printing area is preset according to the classification of direct-to-garment printing fabrics in the printing area of the object to be printed. Based on the color of the direct-to-garment printing block classification, the different direct-to-garment printing fabric classifications, and the fabric thickness, the preset dwell time of the printhead at each printing point in different printing areas is corrected in real time. By adjusting the dwell time of the printhead at each printing point in real time according to the different fabric thicknesses and characteristics of the printing areas, the quality and accuracy of the printing are improved. Real-time correction of the dwell time during the printing process can avoid printing quality problems caused by excessively long or short dwell times, ensuring the consistency and stability of the printing effect.
[0068] In this embodiment, the time for the printhead to stay at each printing point in the printing area is preset according to the classification of direct-to-garment printed fabrics in the printing area of the object to be printed; the dye immersion depth of the printed direct-to-garment printed fabric and the preset dye immersion depth data are sampled and evaluated, and the calculation formula for the preset time for the printhead to stay at each printing point in the printing area is corrected according to the evaluation results.
[0069] By randomly sampling and evaluating the dye penetration depth, the dye penetration depth of the finished direct-to-garment printed fabric is compared with the preset dye penetration depth data. This facilitates the determination that the dye penetration depth meets the preset requirements, thereby improving the quality of the printing effect. Based on the evaluation results, the calculation formula is revised, which helps to improve the accuracy of the preset residence time of the dye dots and avoids excessive adjustments to the printing time of the dye dots in later stages.
[0070] In one embodiment of the present invention, the direct-to-garment printing fabrics are classified as: natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics.
[0071] In one embodiment of the present invention, the time for the nozzle to remain at each printing point in the printing area is preset according to the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, including the following steps:
[0072] Based on the printing area of the object to be printed, direct-print fabrics are classified into natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics.
[0073] The time the printhead spends at each printing point in the printing area is preset using the following formula:
[0074]
[0075] Where T is the time the nozzle stays at each printing point in the printing area, h is the fabric thickness in millimeters, and k is the fabric surface temperature during direct-to-garment printing.
[0076] When the fabric is made of natural animal fiber, α is 0.5 and β is 12; when the fabric is made of natural plant fiber, α is 0.55 and β is 15; when the fabric is made of blended fabric, α is 0.6 and β is 18; when the fabric is made of synthetic fiber, α is 0.6 and β is 20.
[0077] In one embodiment of the present invention, the printed pattern is divided into blocks of different colors, and the blocks of different colors are classified into dark blocks and light blocks. Printing is performed on different blocks, including the following steps:
[0078] The colors of the pattern to be printed are extracted, and the organic plant dyes are adjusted according to the colors extracted from the pattern.
[0079] The Canny edge detection algorithm is used to detect the color change edges on the pattern to be printed at the boundary positions of different color blocks. The color change edge lines are used as the dividing lines of the color blocks to divide the pattern to be printed into different blocks.
[0080] Colors are picked up from different areas of the pattern to be printed, and the colors picked up from different areas are matched with the colors extracted from the pattern to be printed.
[0081] In the CMYK color model, the K channel values of different blocks that match the color are detected. Blocks with K channel values less than a preset threshold are marked as light blocks, and blocks with K channel values greater than or equal to the preset threshold are marked as dark blocks.
[0082] Based on the matched colors, print on the light and dark areas respectively.
[0083] Specifically, in this embodiment, in the CMYK color model, the K channel value of the color matched by different blocks is detected, and blocks with a detected K channel value less than 40 are marked as light blocks, and blocks with a detected K channel value greater than or equal to 40 are marked as dark blocks.
[0084] In one embodiment of the present invention, based on the color of the direct-to-garment printing block classification, the classification of different direct-to-garment printing fabrics, and the thickness of the fabric, the preset value of the dwell time of the printhead at each printing point of the fabric classification in different printing areas is corrected in real time, including the following steps:
[0085] The data acquisition module collects infrared thermal images of the printing area of the object to be printed on fabrics of different thicknesses in different direct-to-garment printing fabric categories during the printing process.
[0086] Organic plant dyes were fluorescently labeled, and the residence time of the dyeing dots and the fluorescence imaging data of the fabrics were collected in different color block categories of fabrics with different thicknesses and different direct-to-garment printing fabrics.
[0087] Train a neural network model based on the data collected by the data acquisition module;
[0088] The system collects infrared thermal imaging images of the printing and dyeing areas in real time, as well as data on direct-to-garment printing fabric classification, fabric thickness, and preset dye immersion depth. Through a trained neural network model, it corrects in real time the preset value of the time the nozzle stays at each printing point in different printing and dyeing areas for each fabric classification.
[0089] In one embodiment of the present invention, the organic plant dye is fluorescently labeled, and the residence time of the dyed dots and the fluorescence imaging data of the fabric are collected in different color block categories of different thicknesses of fabrics in different direct-to-garment printing fabric categories. The process includes the following steps:
[0090] Fluorescent labeling of plant proteins for organic plant dyes;
[0091] Organic plant dyes with fluorescent markings in dark areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics, while organic plant dyes with fluorescent markings in light areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics.
[0092] The fluorescently labeled organic plant dye is excited by an excitation light source to produce fluorescence. A fluorescence microscope is used to image the fabric, capture the fluorescence signal emitted by the fabric, and generate fluorescence imaging data of the fabric surface.
[0093] The dye penetration depth of a portion of the fabric was detected. Using a deep learning algorithm, the organic plant dye penetration depth data of all data was obtained based on the fluorescence imaging data of the fabric surface.
[0094] In one embodiment of the present invention, the dye penetration depth of a portion of the fabric is detected. Using a deep learning algorithm, based on fluorescence imaging data of the fabric surface, the organic plant dye penetration depth data for all data is obtained, including the following steps:
[0095] The depth of dye penetration in some fabrics was detected, and fluorescence imaging data of the fluorescent dyes were obtained. The fluorescence imaging data of dark and light color areas were divided into two groups.
[0096] Preprocessing of fluorescence imaging data includes image resizing, grayscale conversion, and noise removal;
[0097] Convolutional neural network models were trained separately to learn the relationship between dye penetration depth and fluorescence imaging images when the dark and light areas of direct inkjet printing were dyed.
[0098] The fluorescence imaging data of the fabric surface is input into the convolutional neural network model of the corresponding dark and light color blocks of direct inkjet printing, and the dye penetration depth data corresponding to the fluorescence imaging data of the fabric surface is output.
[0099] Specifically, in the CMYK color model of dye colors, the K-channel value represents the amount of black. A large K-channel value in a dye color may affect the fluorescence imaging data of the dye. Fluorescence imaging typically uses ultraviolet light to excite fluorescence, and dyes with higher K-channel values have a greater amount of black, which may block the fluorescence and weaken the detected fluorescence intensity. In this embodiment, the fluorescence imaging data of the fabric surface is input into the convolutional neural network models for the corresponding dark and light areas of the direct-dye printing, and the dye penetration depth data corresponding to all the fluorescence imaging data of the fabric surface is output. The convolutional neural network models for the dark and light areas are trained separately, making the dye penetration depth data obtained from the fluorescence imaging data more accurate.
[0100] In one embodiment of the present invention, training a neural network model based on data collected by the data acquisition module includes the following steps:
[0101] Acquire infrared thermal imaging images of the fabric thickness, the time the nozzle stays at the dyeing point, the dyeing area during dyeing, and data on the dye penetration depth.
[0102] The fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data are used as inputs, and the corresponding nozzle dwell time at the dyeing point is used as the output. CNN neural network models are trained for natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics respectively. These models are used to obtain suggested values for the nozzle dwell time at each dyeing point based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data.
[0103] In one embodiment of the present invention, a preset value for the dwell time of the nozzle at each dyeing point in different dyeing areas of the fabric is corrected in real time using a trained neural network model, including the following steps:
[0104] Based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye immersion depth data, the trained CNN neural network model obtains the suggested value of the nozzle dwell time at each dyeing point.
[0105] Calculate the difference between the preset value of the time the printhead stays at each dyeing point in different dyeing areas and the suggested value of the time the printhead stays at each dyeing point. If the difference is less than the preset difference threshold, the time the printhead stays at each dyeing point will not be modified. Otherwise, the time the printhead stays at each dyeing point will be modified to the suggested value of the time the printhead stays at each dyeing point.
[0106] In one embodiment of the present invention, the dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data are randomly sampled and evaluated. Based on the evaluation results, the calculation formula for the preset time of the printhead staying at each printing point in the printing area is corrected, including the following steps:
[0107] The following formula is used to conduct random sampling and evaluation of the dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data:
[0108]
[0109] Wherein, G is the sampling evaluation value, L1 is the preset dye immersion depth data, and L2 is the dye immersion depth of the sampled direct-to-garment printed fabric.
[0110] If the sampling evaluation value G is greater than 0.3, the coefficient α in the calculation formula for the preset time of the nozzle staying at each printing point in the printing and dyeing area will be reduced by 0.01.
[0111] If the sampling evaluation value G is less than -0.3, the coefficient α in the calculation formula for the preset time for the nozzle to stay at each printing point in the printing and dyeing area will be increased by 0.01.
[0112] Otherwise, the coefficients in the calculation formula for the preset time the nozzle stays at each printing point in the printing area are not adjusted.
[0113] The above embodiments are only used to illustrate the technical methods 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 methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.
Claims
1. A method for direct-to-garment printing with organic plant dyes, characterized in that, Includes the following steps: Import the pattern file to be printed into the image processing module. Based on the printing area of the object to be printed, scale and rotate the pattern to be printed, and position the printing area of the object to be printed, positioning the printed pattern to the printing area of the object to be printed. Input the fabric type of the printing area of the object to be printed, assign the fabric type of the printing area to the direct-to-garment printing fabric category, and detect the fabric thickness. The direct-to-garment printed fabrics are categorized as follows: natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics. Based on the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, the dwell time of the printhead at each printing point in the printing area is preset, including: Based on the classification of direct-to-garment printing fabrics in the printing area of the object to be printed, the dwell time of the printhead at each printing point in the printing area is preset, including the following steps: Based on the printing area of the object to be printed, direct-print fabrics are classified into natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics. The time the printhead spends at each printing point in the printing area is preset using the following formula: Where T is the time the nozzle stays at each printing point in the printing area, h is the fabric thickness in millimeters, and k is the fabric surface temperature during direct-to-garment printing. When the fabric is made of natural animal fiber, α is 0.5 and β is 12; when the fabric is made of natural plant fiber, α is 0.55 and β is 15; when the fabric is made of blended fabric, α is 0.6 and β is 18; when the fabric is made of synthetic fiber, α is 0.6 and β is 20. The printed pattern is divided into blocks of different colors, and these blocks are further classified into dark and light blocks. The different blocks are then printed. Based on the color of the direct-to-garment printing blocks, the classification of different direct-to-garment printing fabrics, and the thickness of the fabrics, the preset values of the dwell time of the print head at each printing point in different printing and dyeing areas are adjusted in real time. The dye immersion depth of the finished direct-to-garment printed fabric and the preset dye immersion depth data are sampled and evaluated. Based on the evaluation results, the calculation formula for the preset time of the print head staying at each printing point in the printing area is revised.
2. The method for direct-to-garment printing of organic plant dyes according to claim 1, characterized in that, The printing pattern is divided into blocks of different colors, and these blocks are further categorized into dark and light color blocks. Printing is then performed on these different blocks, including the following steps: The colors of the pattern to be printed are extracted, and the organic plant dyes are adjusted according to the colors extracted from the pattern. The Canny edge detection algorithm is used to detect the color change edges on the pattern to be printed at the boundary positions of different color blocks. The color change edge lines are used as the dividing lines of the color blocks to divide the pattern to be printed into different blocks. Colors are picked up from different areas of the pattern to be printed, and the colors picked up from different areas are matched with the colors extracted from the pattern to be printed. In the CMYK color model, the K channel values of different blocks that match the color are detected. Blocks with K channel values less than a preset threshold are marked as light blocks, and blocks with K channel values greater than or equal to the preset threshold are marked as dark blocks. Based on the matched colors, print on the light and dark areas respectively.
3. The method for direct-to-garment printing of organic plant dyes according to claim 2, characterized in that, Based on the color of the direct-to-garment printing blocks, the classification of different direct-to-garment printing fabrics, and the fabric thickness, the preset values for the dwell time of the printhead at each printing point in different printing areas of different fabric classifications are adjusted in real time, including the following steps: The data acquisition module collects infrared thermal images of the printing area of the object to be printed on fabrics of different thicknesses in different direct-to-garment printing fabric categories during the printing process. Organic plant dyes were fluorescently labeled, and the residence time of the dyeing dots and the fluorescence imaging data of the fabrics were collected in different color block categories of fabrics with different thicknesses and different direct-to-garment printing fabrics. Train a neural network model based on the data collected by the data acquisition module; The system collects infrared thermal imaging images of the printing and dyeing areas in real time, as well as data on direct-to-garment printing fabric classification, fabric thickness, and preset dye immersion depth. Through a trained neural network model, it corrects in real time the preset value of the time the nozzle stays at each printing point in different printing and dyeing areas for each fabric classification.
4. The method for direct-to-garment printing of organic plant dyes according to claim 3, characterized in that, Fluorescent labeling of organic plant dyes was performed, and fluorescent imaging data of the dyeing point residence time and fabric fluorescence were collected for different color block classifications of fabrics with different thicknesses in different direct-to-garment printing categories. The process included the following steps: Fluorescent labeling of plant proteins for organic plant dyes; Organic plant dyes with fluorescent markings in dark areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics, while organic plant dyes with fluorescent markings in light areas were used to directly print on fabrics of different thicknesses classified as direct-to-garment printing fabrics. The fluorescently labeled organic plant dye is excited by an excitation light source to produce fluorescence. A fluorescence microscope is used to image the fabric, capture the fluorescence signal emitted by the fabric, and generate fluorescence imaging data of the fabric surface. The dye penetration depth of a portion of the fabric was detected. Using a deep learning algorithm, the organic plant dye penetration depth data of all data was obtained based on the fluorescence imaging data of the fabric surface.
5. The method for direct-to-garment printing of organic plant dyes according to claim 4, characterized in that, The dye penetration depth of a portion of the fabric was detected. Using a deep learning algorithm, based on fluorescence imaging data of the fabric surface, the organic plant dye penetration depth data for all data was obtained, including the following steps: The depth of dye penetration in some fabrics was detected, and fluorescence imaging data of the fluorescent dyes were obtained. The fluorescence imaging data of dark and light color areas were divided into two groups. Preprocessing of fluorescence imaging data includes image resizing, grayscale conversion, and noise removal; Convolutional neural network models were trained separately to learn the relationship between dye penetration depth and fluorescence imaging images when the dark and light areas of direct inkjet printing were dyed. The fluorescence imaging data of the fabric surface is input into the convolutional neural network model of the corresponding dark and light color blocks of direct inkjet printing, and the dye penetration depth data corresponding to the fluorescence imaging data of the fabric surface is output.
6. The method for direct-to-garment printing of organic plant dyes according to claim 5, characterized in that, Training a neural network model based on the data collected by the data acquisition module includes the following steps: Acquire infrared thermal imaging images of the fabric thickness, the time the nozzle stays at the dyeing point, the dyeing area during dyeing, and data on the dye penetration depth. The fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data are used as inputs, and the corresponding nozzle dwell time at the dyeing point is used as the output. CNN neural network models are trained for natural animal fiber fabrics, natural plant fiber fabrics, synthetic fiber fabrics, and blended fabrics respectively. These models are used to obtain suggested values for the nozzle dwell time at each dyeing point based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye penetration depth data.
7. The method for direct-to-garment printing of organic plant dyes according to claim 6, characterized in that, The preset time for the nozzle to remain at each dyeing point in different fabric categories in different dyeing areas is corrected in real time using a trained neural network model, including the following steps: Based on the fabric thickness, infrared thermal imaging images of the dyeing area during dyeing, and dye immersion depth data, the trained CNN neural network model obtains the suggested value of the nozzle dwell time at each dyeing point. Calculate the difference between the preset value of the time the printhead stays at each dyeing point in different dyeing areas and the suggested value of the time the printhead stays at each dyeing point. If the difference is less than the preset difference threshold, the time the printhead stays at each dyeing point will not be modified. Otherwise, the time the printhead stays at each dyeing point will be modified to the suggested value of the time the printhead stays at each dyeing point.
8. The method for direct-to-garment printing of organic plant dyes according to claim 7, characterized in that, The dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data are randomly sampled and evaluated. Based on the evaluation results, the calculation formula for the preset time of the printhead's residence at each printing point in the printing area is revised, including the following steps: The following formula is used to conduct random sampling and evaluation of the dye penetration depth of the finished direct-to-garment printed fabric and the preset dye penetration depth data: Wherein, G is the sampling evaluation value, L1 is the preset dye immersion depth data, and L2 is the dye immersion depth of the sampled direct-to-garment printed fabric. If the sampling evaluation value G is greater than 0.3, the coefficient α in the calculation formula for the preset time of the nozzle staying at each printing point in the printing and dyeing area will be reduced by 0.
01. If the sampling evaluation value G is less than -0.3, the coefficient α in the calculation formula for the preset time for the nozzle to stay at each printing point in the printing and dyeing area will be increased by 0.
01. Otherwise, the coefficients in the calculation formula for the preset time the nozzle stays at each printing point in the printing area are not adjusted.