Flannel non-watermark printing intelligent control method and system

Abstract

The application provides a flannel watermark-free printing intelligent control method and system, relates to the technical field of fabric printing, and comprises the following steps: obtaining fiber orientation distribution data and target printing pattern data of flannel fabric to be printed, calculating a fiber orientation vector and a permeability coefficient, direction-adaptively decomposing the target printing pattern to obtain a sub-pattern unit, generating a printing application track, determining an application pressure value and an application rate value of each track segment to generate a control parameter sequence, driving a printing device according to the control parameter sequence during execution, collecting deformation response data in real time and constructing a deformation-fiber deviation field to identify a deviation propagation path, and predictively compensating the control parameter sequence of the unexecuted track segment, so that the anisotropic structure of the flannel fabric is adaptively printed, and the printing precision and consistency are improved.

Description

Technical Field

[0001] This invention relates to the field of fabric printing technology, and in particular to an intelligent control method and system for waterless printing of flannel. Background Technology

[0002] In the field of waterless printing technology for flannel, existing printing control methods are typically based on fixed process parameters. The printing device moves at a constant speed along a preset two-dimensional trajectory and applies ink or dye to the fabric surface with constant pressure. This conventional approach assumes that the fabric is isotropic and ignores the differences in permeability caused by the different fiber orientation distributions in flannel fabric. The generation of the printed pattern only considers the geometry and color distribution of the target pattern, without incorporating the fiber orientation vectors on the fabric surface into the planning of the printing trajectory and process parameters.

[0003] Conventional methods have several drawbacks, making it difficult to guarantee uniform pigment penetration and clear pattern boundaries. The fiber orientation of flannel fabric varies, resulting in significant anisotropy in capillary penetration capacity for printing in different areas. Applying uniform pressure and rate leads to excessive pigment diffusion along the fiber orientation direction and insufficient penetration in the vertical direction, ultimately causing blurred pattern edges and color bleeding between color blocks, affecting print quality. Even with subsequent processing adjustments, it is difficult to completely eliminate this uneven color distribution caused by differences in fiber orientation.

[0004] Other shortcomings include insufficient real-time adaptability to fabric deformation during the printing process. After setting fixed parameters before application, the system operates in an open-loop manner, failing to respond to localized deformations caused by stress or humidity changes in the fabric. During actual processing, flannel fabrics may stretch, shrink, or twist, causing a deviation between the predetermined printing trajectory and fiber orientation. This dynamic deviation cannot be detected or corrected, resulting in the actual application position deviating from the target area, further exacerbating pattern distortion and uneven penetration. The lack of methods to predict and compensate for short-term changes in fiber orientation caused by deformation makes it difficult to meet the requirements of automated production in terms of printing accuracy and yield. Summary of the Invention

[0005] The present invention provides an intelligent control method and system for waterless printing of flannel, which can solve the problems in the prior art.

[0006] A first aspect of the present invention provides an intelligent control method for waterless printing of flannel, comprising: Obtain fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; Calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data, and perform directional adaptive decomposition on the target printed pattern data based on the fiber orientation vector and penetration coefficient to obtain multiple sub-pattern units corresponding to different fiber orientation regions. Based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area, a printing application trajectory is generated; Based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, the application pressure value and application rate value of each trajectory segment are determined, and a control parameter sequence is generated. In the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence. The deformation response data of the fabric is collected in real time. Based on the deformation response data, a deformation-fiber deviation field is constructed and the deviation propagation path is identified. Predictive compensation is performed on the control parameter sequence of the unexecuted trajectory segment.

[0007] In one optional embodiment, the fiber orientation vector and penetration coefficient of each region on the fabric surface are calculated based on fiber orientation distribution data. The target printed pattern data is then subjected to directional adaptive decomposition based on the fiber orientation vector and penetration coefficient to obtain multiple sub-pattern units corresponding to different fiber orientation regions, including: Extract fiber orientation angle and fiber density data from each sampling point in the fiber orientation distribution data. Construct a fiber orientation vector field for each region based on the fiber orientation angle. Calculate the anisotropic permeability coefficient and directional permeability resistance coefficient for each region based on the fiber density data, the spatial gradient of the fiber orientation vector field, and the angle between fiber orientation vectors of adjacent regions. Extract the pattern edge direction distribution and the internal texture direction distribution of the target printed pattern data, and construct the direction feature tensor field; Spatially register the orientation feature tensor field with the fiber orientation vector field, calculate the orientation fit index at each spatial location, and determine the decomposition boundary position and decomposition hierarchy structure of the target printed pattern data based on the spatial distribution characteristics of the orientation fit index. The target printed pattern data is decomposed along the decomposition boundary to obtain multiple sub-pattern units. The pattern orientation features of each sub-pattern unit are coordinated with the fiber orientation vector and permeability coefficient distribution characteristics of the corresponding fiber orientation region. Furthermore, at the decomposition boundary, the pattern orientation features of adjacent sub-pattern units match the fiber orientation vector conversion angle at the corresponding position.

[0008] In one optional embodiment, spatial registration is performed between the orientation feature tensor field and the fiber orientation vector field, and the orientation fit index at each spatial location is calculated. The decomposition boundary position and decomposition hierarchy of the target printed pattern data are determined based on the spatial distribution characteristics of the orientation fit index, including: Spatial alignment and registration are performed on the orientation feature tensor field and the fiber orientation vector field; Extract the principal direction components of the orientation feature tensor field at each spatial location, calculate the angle deviation between the principal direction components and the corresponding fiber orientation vector, and calculate the orientation coupling strength at each spatial location based on the orientation dispersion of the orientation feature tensor field at each spatial location and the spatial gradient of the fiber orientation vector field. Based on the included angle deviation and the directional coupling strength, and combined with the anisotropic permeability coefficient at the corresponding position, the directional fit index of each spatial position is calculated. Identify the spatial gradient abrupt change locations of the orientation fitness index and the connected regions where the orientation fitness index is lower than a preset fitness threshold. Determine the spatial gradient abrupt change locations as the first-level decomposition boundaries and the boundaries of the connected regions as the second-level decomposition boundaries. Based on the spatial topological relationship between the first-level decomposition boundary and the second-level decomposition boundary, a decomposition hierarchy structure containing a first decomposition level and a second decomposition level is constructed, wherein the first decomposition level corresponds to the region divided by the first-level decomposition boundary, and the second decomposition level corresponds to the refined division region of the second-level decomposition boundary within the first decomposition level.

[0009] In one optional embodiment, generating the printing application trajectory based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding region includes: The spatial distribution and boundary contours of multiple sub-pattern units are obtained, and the fiber orientation vector field and anisotropic permeability coefficient distribution of the corresponding region of each sub-pattern unit are extracted. Based on the boundary contour of each sub-pattern unit and the fiber orientation vector field of the corresponding region, the trajectory coverage and trajectory starting point of each sub-pattern unit are determined, and the trajectory density distribution and trajectory spacing inside each sub-pattern unit are calculated based on the spatial gradient and anisotropic penetration coefficient of the fiber orientation vector field. Based on the trajectory density distribution and trajectory spacing, multiple trajectory segments are generated along the fiber orientation vector direction of the corresponding region of each sub-pattern unit; Extract the fiber orientation vector and anisotropic penetration coefficient at the boundary of adjacent sub-pattern units. Generate a trajectory transition segment based on the angular difference between the fiber orientation vectors of adjacent regions at the boundary and the anisotropic penetration coefficient. Connect the trajectory transition segment to the trajectory segment of adjacent sub-pattern units to form a complete printing application trajectory.

[0010] In one optional embodiment, generating multiple trajectory segments along the fiber orientation vector direction of the corresponding region of each sub-pattern unit, based on the trajectory density distribution and trajectory spacing, includes: Based on the trajectory density distribution and trajectory spacing, multiple trajectory starting points are set within each sub-pattern unit; The corresponding region of each sub-pattern unit is divided into multiple grid units, and the fiber orientation vector and anisotropic permeability coefficient of each grid unit are extracted. Starting from the starting point of each trajectory, read the fiber orientation vector and anisotropic permeability coefficient of the current grid cell, determine the direction of movement based on the fiber orientation vector, determine the step length based on the anisotropic permeability coefficient, and move the step length along the direction of movement to reach the next grid cell. Establish a local coordinate system at each grid cell, take the fiber orientation vector as the principal axis direction of the local coordinate system, calculate the rotation angle between the local coordinate systems of adjacent grid cells, and sum the rotation angles to obtain the cumulative deflection of the trajectory path. When the cumulative deflection exceeds the preset deflection threshold, multiple candidate grid cells are searched around the current grid cell. The similarity between the fiber orientation vector of each candidate grid cell and the fiber orientation vector of the starting grid cell is calculated, and the candidate grid cell with the highest similarity is selected as the regression target point. Move from the current grid cell toward the target point of regression, clear the accumulated deflection to zero, and then continue moving along the fiber orientation vector direction; The coordinates of each grid cell passed during the forward movement are recorded to form a position sequence, generating a trajectory segment.

[0011] In one optional embodiment, based on the anisotropic penetration coefficient and fiber orientation vector at corresponding positions of each trajectory segment in the printing application trajectory, the applied pressure value and applied rate value of each trajectory segment are determined, and a control parameter sequence is generated, including: Extract the anisotropic penetration coefficients at corresponding positions of each trajectory segment in the printing application trajectory, construct a penetration coefficient distribution curve, calculate the curvature value of the penetration coefficient distribution curve, and identify trajectory segments whose penetration coefficient changes exceed a preset curvature threshold as target control trajectory segments. The anisotropic permeability coefficient of the target control trajectory segment is divided into intervals, and the anisotropic permeability coefficient of each interval is mapped to the corresponding applied pressure adjustment factor. The applied pressure adjustment factor is weighted and calculated with the preset benchmark pressure value to obtain the applied pressure value of each trajectory segment. Extract the fiber orientation vectors at corresponding positions of adjacent trajectory segments in the printing application trajectory, calculate the angle between adjacent fiber orientation vectors to obtain the fiber turning angle sequence, and perform differential operation on the fiber turning angle sequence to obtain the fiber turning angular velocity; The applied rate compensation amount is determined based on the fiber turning angular velocity, and the applied rate compensation amount is superimposed on the preset reference rate value to obtain the applied rate value for each trajectory segment. The applied pressure value and application rate value are associated and bound with the position coordinates of the corresponding trajectory segment, and a sequence of control parameters is generated according to the trajectory segment order of the printing application trajectory.

[0012] In one optional embodiment, during the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence, and the deformation response data of the fabric is collected in real time. A deformation-fiber deviation field is constructed based on the deformation response data, and the deviation propagation path is identified. Predictive compensation for the control parameter sequence of the unexecuted trajectory segment includes: The printing device is driven to perform the application operation of each trajectory segment in sequence along the printing application trajectory according to the control parameter sequence. During the execution, the deformation response data of the fabric surface is collected in real time by the deformation sensor and the deformation response data is converted into a deformation response vector field. The deformation response vector field is spatially superimposed with the fiber orientation distribution data at the corresponding position of the executed trajectory segment to construct a deformation-fiber deviation field. The spatial distribution gradient of the deviation vector in the deformation-fiber deviation field is extracted, and the deviation propagation path is identified based on the spatial distribution gradient. Tracing back along the deviation propagation path to the unexecuted trajectory segment, the coupling coefficient between the deviation vector at each position point on the deviation propagation path and the fiber orientation vector corresponding to each position point is calculated. Based on the coupling coefficient, predictive compensation is performed on the applied pressure value and applied rate value of the unexecuted trajectory segment. The compensated applied pressure and applied rate values ​​are updated to the control parameters of the corresponding unexecuted trajectory segment in the control parameter sequence, and the printing device continues to be driven to perform the application operation of the remaining trajectory segment according to the updated control parameter sequence.

[0013] A second aspect of the present invention provides an intelligent control system for waterless printing of flannel, comprising: The data acquisition unit is used to acquire fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; The pattern decomposition unit is used to calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data. Based on the fiber orientation vector and penetration coefficient, the target printed pattern data is decomposed in an adaptive direction to obtain multiple sub-pattern units corresponding to different fiber orientation regions. The trajectory generation unit is used to generate a printing application trajectory based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area; The parameter determination unit is used to determine the application pressure and application rate values ​​of each trajectory segment based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, and to generate a sequence of control parameters. The compensation control unit is used to drive the printing device to apply the printing along the printing application trajectory according to the control parameter sequence during the waterless printing operation, collect the deformation response data of the fabric in real time, construct the deformation-fiber deviation field based on the deformation response data and identify the deviation propagation path, and perform predictive compensation for the control parameter sequence of the unexecuted trajectory segment.

[0014] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0016] In this embodiment of the invention, the directional adaptive decomposition based on fiber orientation distribution and penetration coefficient enables precise matching between the printed pattern and the anisotropic structure of the fabric. Sub-pattern units are applied according to the fiber orientation vector, avoiding bleeding, smudging, or uneven color caused by differences in fiber orientation. The applied pressure and application rate values ​​for each trajectory segment are independently controlled according to the anisotropic penetration coefficient, achieving uniform penetration of dyes or pigments within the fabric and significantly improving the clarity of pattern edges and color saturation. Deformation response data is collected in real time and a deformation-fiber deviation field is constructed to dynamically identify the propagation path of deviations in fiber orientation. Predictive compensation is performed on unexecuted trajectory segments, adjusting the application parameters in advance to counteract the cumulative effect of deformation. This significantly reduces pattern distortion or misalignment caused by fabric stretching, wrinkling, or rebound, and is particularly suitable for highly elastic or easily deformable flannel materials. The entire control method eliminates the need for water washing or steam fixing, reducing water consumption and wastewater discharge. The intelligent compensation mechanism makes the printing process highly robust to batch differences in raw materials and environmental changes, reducing the scrap rate. The color reproduction accuracy and batch consistency are significantly better than traditional waterless printing processes, meeting the quality requirements of high-end flannel products. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the intelligent control method for waterless printing of flannel; Figure 2 This is a flowchart of the adaptive compensation control logic for the printing device. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0019] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0020] Figure 1 This is a flowchart illustrating the intelligent control method for waterless printing of flannel according to an embodiment of the present invention. Figure 1 As shown, the method includes: Obtain fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; Calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data, and perform directional adaptive decomposition on the target printed pattern data based on the fiber orientation vector and penetration coefficient to obtain multiple sub-pattern units corresponding to different fiber orientation regions. Based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area, a printing application trajectory is generated; Based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, the application pressure value and application rate value of each trajectory segment are determined, and a control parameter sequence is generated. In the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence. The deformation response data of the fabric is collected in real time. Based on the deformation response data, a deformation-fiber deviation field is constructed and the deviation propagation path is identified. Predictive compensation is performed on the control parameter sequence of the unexecuted trajectory segment.

[0021] In one optional embodiment, the fiber orientation vector and penetration coefficient of each region on the fabric surface are calculated based on fiber orientation distribution data. The target printed pattern data is then subjected to directional adaptive decomposition based on the fiber orientation vector and penetration coefficient to obtain multiple sub-pattern units corresponding to different fiber orientation regions, including: Extract fiber orientation angle and fiber density data from each sampling point in the fiber orientation distribution data. Construct a fiber orientation vector field for each region based on the fiber orientation angle. Calculate the anisotropic permeability coefficient and directional permeability resistance coefficient for each region based on the fiber density data, the spatial gradient of the fiber orientation vector field, and the angle between fiber orientation vectors of adjacent regions. Extract the pattern edge direction distribution and the internal texture direction distribution of the target printed pattern data, and construct the direction feature tensor field; Spatially register the orientation feature tensor field with the fiber orientation vector field, calculate the orientation fit index at each spatial location, and determine the decomposition boundary position and decomposition hierarchy structure of the target printed pattern data based on the spatial distribution characteristics of the orientation fit index. The target printed pattern data is decomposed along the decomposition boundary to obtain multiple sub-pattern units. The pattern orientation features of each sub-pattern unit are coordinated with the fiber orientation vector and permeability coefficient distribution characteristics of the corresponding fiber orientation region. Furthermore, at the decomposition boundary, the pattern orientation features of adjacent sub-pattern units match the fiber orientation vector conversion angle at the corresponding position.

[0022] In one specific implementation, after acquiring the fiber orientation distribution data of the flannel fabric to be printed, a refined fiber orientation analysis of the fabric surface is required. The fabric surface is scanned using a high-resolution image acquisition device, and a two-dimensional coordinate system is established on the fabric surface, dividing it into several sampling grid units, each with a size of 0.5 mm × 0.5 mm. At each sampling point, the fiber orientation angle at that location is obtained using polarized light microscopy, defined as the angle between the fiber principal axis and the horizontal axis of the coordinate system, ranging from 0 degrees to 180 degrees. Simultaneously, the fiber density data at that sampling point is obtained using image grayscale analysis technology. Fiber density is quantified by the number of fibers per unit area, expressed as fibers per square millimeter.

[0023] Based on the fiber orientation angle data at each sampling point, a fiber orientation vector field is constructed on the fabric surface. For coordinate positions... The sampling point, whose fiber direction angle is denoted as Then the fiber orientation vector at that position is represented as Spatial interpolation of the fiber orientation vectors at all sampling points is performed using a bicubic spline interpolation method to obtain a continuous fiber orientation vector field on the fabric surface. This vector field can reflect the fiber orientation characteristics at any position on the fabric surface, providing basic data support for subsequent permeability coefficient calculation and pattern decomposition.

[0024] After constructing the fiber orientation vector field, it is necessary to calculate the anisotropic penetration coefficient of each region. The penetration coefficient reflects the ability of printing dyes to penetrate and diffuse into the interfiber spaces of the fabric. This coefficient is influenced by the combined effects of fiber density, fiber alignment, and differences in fiber orientation between adjacent regions. For a specific region on the fabric surface, the average fiber density of that region is extracted. Higher fiber density and smaller fiber gaps result in greater permeability resistance. The spatial gradient of the fiber orientation vector field in this region is calculated; the gradient magnitude reflects the degree of change in fiber orientation. The gradient value is obtained by performing partial derivative calculations on the fiber orientation vector field. Simultaneously, the angle between this region and adjacent regions' fiber orientation vectors is analyzed. A larger angle indicates a significant change in fiber orientation at the region boundary, potentially creating a permeability barrier effect.

[0025] The calculation of anisotropic permeability coefficient takes into account the above three factors. Permeability coefficient along the main fiber orientation direction. It is inversely proportional to fiber density and is also affected by the regularity of fiber arrangement. The permeability coefficient is perpendicular to the main fiber orientation direction. The directional permeability resistance coefficient is typically smaller than that in the parallel direction, and the ratio of the two reflects the degree of anisotropy in permeability. This coefficient is used to quantify the hindering effect of fiber orientation changes on the penetration process. It is related to the spatial gradient of the fiber orientation vector field and the angle between fiber orientation vectors in adjacent regions. When the angle between fiber orientation vectors in adjacent regions exceeds 30 degrees, the directional penetration resistance coefficient increases significantly, indicating that dye penetration will be significantly hindered at this location, requiring targeted adjustments during the printing process.

[0026] After completing the fabric fiber characteristic analysis, directional feature extraction is required for the target printed pattern data. The target printed pattern data is stored in digital image format, containing color, texture, and geometric information. Edge contours of the pattern are extracted using an edge detection algorithm, and the Canny edge detection operator is used to identify the position and direction of the pattern edges. For each detected edge segment, its tangent direction is calculated, representing the local direction of the pattern edge. Statistical analysis of the directional information of all edge segments yields the pattern edge directional distribution characteristics. For the internal region of the pattern, a Gabor filter bank is used to perform multi-directional analysis of the pattern texture, extracting the dominant direction and directional distribution range of the internal texture.

[0027] Based on the extracted pattern edge orientation distribution and pattern internal texture orientation distribution, an orientation feature tensor field is constructed. For each pixel position in the pattern... Calculate the orientation feature tensor at this location. This tensor is a second-order symmetric tensor, whose eigenvalues ​​and eigenvectors reflect the intensity and principal direction of the pattern's directional features at that location, respectively. Eigenvectors with larger eigenvalues ​​correspond to the principal direction of the pattern, while eigenvectors with smaller eigenvalues ​​correspond to the secondary direction. By performing tensor calculations on the entire pattern region, a complete directional feature tensor field is obtained, which can comprehensively describe the directional characteristics of the target printed pattern at various locations in space.

[0028] Spatial registration of the orientation feature tensor field with the fiber orientation vector field is a crucial step in achieving orientation adaptive decomposition. Since the two fields may differ in spatial resolution and coordinate system, coordinate transformation and scale normalization are first required to ensure alignment within the same spatial reference frame. After registration, the fiber orientation vector at each spatial location on the fabric surface is extracted. and pattern orientation feature tensor Main eigenvectors Calculate the angle between two direction vectors. This included angle reflects the degree of matching between the pattern direction and the fiber orientation.

[0029] Orientation Adaptability Index This index is used to quantify the synergy between pattern orientation and fiber orientation. When the pattern orientation is parallel or nearly parallel to the fiber orientation, the dye penetrates and diffuses along the fiber direction, resulting in a clear printing effect with sharp edges; in this case, the orientation fit index takes a higher value. When the pattern orientation is perpendicular to the fiber orientation or at a large angle, dye penetration is hindered by the fiber arrangement, easily leading to uneven penetration or blurred edges; in this case, the orientation fit index takes a lower value. The orientation fit index also needs to consider the penetration coefficient at that location. In areas with a low penetration coefficient, even with a high orientation fit, dye penetration is still limited; therefore, the orientation fit index needs to be corrected.

[0030] By analyzing the spatial distribution characteristics of the orientation fit index, locations where significant changes in orientation fit are identified; these locations serve as candidate decomposition boundaries for the target printed pattern data. Regions with large spatial gradients in the orientation fit index indicate significant changes in pattern orientation and fiber orientation characteristics, making them suitable as decomposition boundaries. Simultaneously, considering the geometric structure of the pattern itself, natural boundaries are preferentially chosen as decomposition boundaries to avoid forced segmentation within continuous areas of the pattern. Determining the decomposition hierarchy requires comprehensive consideration of both pattern complexity and the degree of fiber orientation variation. For fabric areas with drastic fiber orientation changes, finer decomposition and increased decomposition levels are necessary; for areas with relatively uniform fiber orientation, a coarser decomposition granularity can be used.

[0031] The target printed pattern data is decomposed along the defined decomposition boundary. The decomposition process employs a region growing algorithm, starting from a seed point with a high orientation fit index and gradually expanding outwards, merging spatially contiguous pixels with similar orientation fit indices into the same sub-pattern unit. During region growing, a threshold range for the orientation fit index is set; when the difference in orientation fit indices between adjacent pixels exceeds the threshold, expansion stops and is marked as a decomposition boundary. Through iterative execution of the region growing process, the entire target printed pattern is ultimately decomposed into multiple sub-pattern units, each corresponding to a specific fiber orientation region on the fabric surface.

[0032] The pattern orientation characteristics of each sub-pattern unit are coordinated with the fiber orientation vector and penetration coefficient distribution characteristics of the corresponding fiber orientation region. For each sub-pattern unit, the pattern orientation within it is mainly distributed along the main fiber orientation direction of that region, and the angle between the pattern edge direction and the fiber arrangement direction is controlled within 15 degrees. The color depth and texture density of the sub-pattern unit are adjusted according to the penetration coefficient of the corresponding region. In areas with a higher penetration coefficient, higher color saturation and denser texture can be used; in areas with a lower penetration coefficient, it is necessary to reduce color saturation or texture density to avoid excessive dye accumulation that would lead to a decrease in printing quality.

[0033] At the decomposition boundary, a smooth transition is required between adjacent sub-pattern units. The fiber orientation vector at the decomposition boundary is extracted, and the conversion angle between the fiber orientation vectors of adjacent sub-pattern units on either side of this boundary is calculated. The pattern orientation features of adjacent sub-pattern units are gradually adjusted at the boundary to match the change in pattern orientation with the conversion angle of the fiber orientation vector. Specifically, a transition band is set at the decomposition boundary, and its width is adaptively adjusted according to the magnitude of the fiber orientation conversion angle; the larger the conversion angle, the wider the transition band. Within the transition band, the pattern orientation features, color intensity, and texture density are linearly or non-linearly interpolated to ensure visual and physical continuity between adjacent sub-pattern units, avoiding obvious printing breaks or color differences at the decomposition boundary.

[0034] In one optional embodiment, spatial registration is performed between the orientation feature tensor field and the fiber orientation vector field, and the orientation fit index at each spatial location is calculated. The decomposition boundary position and decomposition hierarchy of the target printed pattern data are determined based on the spatial distribution characteristics of the orientation fit index, including: Spatial alignment and registration are performed on the orientation feature tensor field and the fiber orientation vector field; Extract the principal direction components of the orientation feature tensor field at each spatial location, calculate the angle deviation between the principal direction components and the corresponding fiber orientation vector, and calculate the orientation coupling strength at each spatial location based on the orientation dispersion of the orientation feature tensor field at each spatial location and the spatial gradient of the fiber orientation vector field. Based on the included angle deviation and the directional coupling strength, and combined with the anisotropic permeability coefficient at the corresponding position, the directional fit index of each spatial position is calculated. Identify the spatial gradient abrupt change locations of the orientation fitness index and the connected regions where the orientation fitness index is lower than a preset fitness threshold. Determine the spatial gradient abrupt change locations as first-level decomposition boundaries and the boundaries of the connected regions as second-level decomposition boundaries. Based on the spatial topological relationship between the first-level decomposition boundary and the second-level decomposition boundary, a decomposition hierarchy structure containing a first decomposition level and a second decomposition level is constructed, wherein the first decomposition level corresponds to the region divided by the first-level decomposition boundary, and the second decomposition level corresponds to the refined division region of the second-level decomposition boundary within the first decomposition level.

[0035] In one specific implementation, after constructing the fiber orientation vector field and the orientation feature tensor field, spatial registration is required to establish an accurate correspondence. The registration process employs an affine transformation method based on feature points. First, several marker points are selected on the fabric surface as references. The positions of these marker points in the physical coordinate system are acquired using an image acquisition device, and their corresponding positions in the pattern data coordinate system are recorded. The affine transformation matrix is ​​solved using the least squares method, mapping the orientation feature tensor field in the pattern coordinate system to the fabric physical coordinate system, ensuring that the pattern orientation information at each spatial location is within the same coordinate frame as the fiber orientation information at that location. After registration, bilinear interpolation is performed on the transformed orientation feature tensor field to ensure its spatial resolution remains consistent with the fiber orientation vector field, thus providing an accurate data foundation for subsequent orientation fit calculations.

[0036] After spatial registration, the principal directional components of the directional feature tensor field at each spatial location are extracted. The directional feature tensor at each location can be represented as a second-order symmetric tensor, and its principal eigenvector is obtained through eigenvalue decomposition. This principal eigenvector represents the main directional feature of the pattern at that location. The angular deviation between the principal eigenvector and the corresponding fiber orientation vector is calculated. This angular deviation reflects the degree of consistency between the pattern orientation and the fiber orientation. The smaller the angular deviation, the closer the pattern orientation is to the fiber orientation, and the more consistent the dye penetration behavior along the fiber direction during printing is with the pattern design intent. To quantify this deviation, an angular deviation index is introduced. The calculation method is the absolute value of the cosine of the angle between the main feature vector and the fiber orientation vector. The value ranges from 0 to 1. The closer the value is to 1, the better the directional consistency.

[0037] The calculation of directional coupling strength requires comprehensive consideration of the directional dispersion of the directional characteristic tensor field and the spatial gradient of the fiber orientation vector field. The directional dispersion of the directional characteristic tensor field is characterized by calculating the anisotropy index of the tensor, which is defined as the ratio of the difference between the largest and smallest eigenvalues ​​of the tensor to their sum, denoted as [equation missing]. .when When it is close to 1, it indicates that the pattern at that location has obvious directional characteristics; when When the value is close to 0, it indicates that the pattern orientation characteristics at that location are not obvious or are isotropic. The spatial gradient of the fiber orientation vector field is obtained by calculating the rate of change of the orientation vectors of adjacent locations, denoted as . Its magnitude reflects the degree of drastic change in fiber orientation in space. Orientation coupling strength. Combining these two factors, it can be expressed as and The weighted combination is adjusted according to the actual fabric type and printing process requirements. High directional coupling strength means that the position has both obvious pattern direction characteristics and is in a region with large changes in fiber orientation. Such positions are more sensitive to applied parameters during the printing process and require more refined control strategies.

[0038] The orientation fit index integrates information from three dimensions: angular deviation, directional coupling strength, and anisotropic penetration coefficient. The anisotropic penetration coefficient reflects the difference in dye penetration ability in different directions. When the pattern direction aligns with the fiber orientation, a larger anisotropic penetration coefficient at that location allows the dye to diffuse more easily along the expected direction, resulting in a more ideal printing effect. Orientation Fit Index The calculation formula is: ,in The normalized anisotropic penetration coefficient is obtained by dividing the anisotropic penetration coefficient by the global maximum value. This index comprehensively reflects the overall performance of a spatial location in terms of directional consistency, directional characteristic significance, and controllability of penetration behavior. The higher the value, the more suitable the location is for printing according to the current pattern design.

[0039] After obtaining the global orientation fit index distribution, abrupt changes in orientation fit are identified through spatial gradient analysis. The Sobel operator or Canny edge detection algorithm is used to calculate the spatial gradient magnitude of the orientation fit index. When the gradient magnitude exceeds a preset gradient threshold, the location is considered to have a significant change in orientation fit and is marked as a spatial gradient abrupt change location. These abrupt change locations typically correspond to the boundaries of different directional feature regions in the pattern, or areas where fiber orientation changes drastically. Connecting all identified spatial gradient abrupt change locations forms closed or semi-closed boundary curves, which constitute the first-level decomposition boundaries. The first-level decomposition boundaries divide the entire printing area into several large sub-regions. The orientation fit within each sub-region is relatively stable, while the directional feature differences between regions are significant.

[0040] In addition to identifying gradient abrupt changes, it is also necessary to identify connected regions where the orientation fit index is below a preset fit threshold. This can be achieved by setting a fit threshold. The distribution of the orientation fit index is binarized. The locations are marked as low-fit regions. A connected component analysis algorithm is used to identify the connected components within all low-fit regions. Each connected component represents a region with poor orientation fit. These regions may be due to a severe mismatch between the pattern orientation and fiber orientation, or extremely uneven fiber orientation in the region, making penetration behavior difficult to predict. The boundary of each connected region is extracted as the second-level decomposition boundary. The second-level decomposition boundary is typically located within the sub-regions defined by the first-level decomposition boundary, used to further refine regions that still have orientation fit problems.

[0041] Constructing a hierarchical decomposition structure requires analyzing the spatial topological relationship between the first-level and second-level decomposition boundaries. First, the printing area is divided into multiple first-level decomposition regions based on the first-level decomposition boundaries, each with relatively consistent directional adaptation characteristics. Then, it is checked whether a second-level decomposition boundary exists within each first-level region. If so, multiple second-level decomposition regions are further subdivided within that first-level region. These second-level regions refine the first-level regions, specifically addressing cases where overall directional adaptation is acceptable but local low-fit connectivity exists. This two-level decomposition structure ensures both consistent directional division at a large scale and optimization of local details. Each decomposition region is assigned a unique identifier, and its level, boundary coordinates, internal average directional adaptation, and corresponding fiber orientation statistical characteristics are recorded. This information will be used for the generation of subsequent sub-pattern units and the planning of the printing trajectory.

[0042] In practice, the first decomposition level primarily addresses directional conflicts between different pattern elements. For example, a printed pattern may contain horizontal and vertical stripes with significantly different directional characteristics. The first-level decomposition boundary separates them into different sub-regions for individual directional adaptation. The second decomposition level addresses localized inhomogeneities within a single pattern element. For instance, if fiber orientation in a certain area becomes disordered due to weaving process fluctuations, resulting in a significantly lower directional fit compared to surrounding areas, the second-level decomposition boundary isolates this localized area and applies a special compensation strategy. This layered decomposition strategy allows the printing control system to flexibly address directional adaptation issues at different scales, ensuring both the integrity of the overall printed pattern and precise control of local details.

[0043] The construction of the decomposition hierarchy also needs to consider the smoothness and continuity of the boundaries. Since boundaries obtained directly through gradient abrupt changes and threshold segmentation may be jagged or discontinuous, they need to be smoothed. Spline curve fitting or morphological closing operations are used to optimize the boundaries, ensuring that the boundary curves are smooth and closed, avoiding path discontinuities or overlaps in subsequent trajectory generation. Simultaneously, for decomposition regions with excessively small areas, they can be merged according to a preset minimum area threshold, incorporating them into adjacent regions with the closest directional fit, thereby reducing the complexity of the decomposition hierarchy and improving printing efficiency.

[0044] In one optional embodiment, generating the printing application trajectory based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding region includes: The spatial distribution and boundary contours of multiple sub-pattern units are obtained, and the fiber orientation vector field and anisotropic permeability coefficient distribution of the corresponding region of each sub-pattern unit are extracted. Based on the boundary contour of each sub-pattern unit and the fiber orientation vector field of the corresponding region, the trajectory coverage and trajectory starting point of each sub-pattern unit are determined, and the trajectory density distribution and trajectory spacing inside each sub-pattern unit are calculated based on the spatial gradient and anisotropic penetration coefficient of the fiber orientation vector field. Based on the trajectory density distribution and trajectory spacing, multiple trajectory segments are generated along the fiber orientation vector direction of the corresponding region of each sub-pattern unit; Extract the fiber orientation vector and anisotropic penetration coefficient at the boundary of adjacent sub-pattern units. Generate a trajectory transition segment based on the angular difference between the fiber orientation vectors of adjacent regions at the boundary and the anisotropic penetration coefficient. Connect the trajectory transition segment to the trajectory segment of adjacent sub-pattern units to form a complete printing application trajectory.

[0045] In one specific implementation, after obtaining multiple sub-pattern units, it is necessary to determine the spatial distribution position of each sub-pattern unit on the fabric surface. Image registration technology is used to map the pixel coordinates of the sub-pattern units to the physical coordinate system of the fabric surface, recording the center coordinates, the bounding rectangle range, and the actual set of pixels occupied by each sub-pattern unit. For each sub-pattern unit, a boundary tracing algorithm is used to extract its boundary contour. This algorithm starts from the top-left pixel of the sub-pattern unit and searches neighboring pixels sequentially in a clockwise direction, determining whether a pixel belongs to that sub-pattern unit. If it does, the tracing continues; otherwise, it is recorded as a boundary point. In this way, a closed contour curve composed of a series of boundary points can be obtained, which accurately describes the shape characteristics of the sub-pattern unit.

[0046] While extracting the boundary contour, it is necessary to obtain the fiber orientation vector field of the corresponding region for each sub-pattern unit. Based on the fiber orientation vector field of the fabric surface established in the previous steps, the fiber orientation vector within the coverage area of ​​each sub-pattern unit is quickly queried using a spatial indexing method. For any position within a sub-pattern unit, its fiber orientation vector can be calculated from the fiber orientation vectors of the surrounding four sampling points using bilinear interpolation. Simultaneously, the anisotropic permeability coefficient distribution of this region is extracted. This distribution reflects the difference in the fabric's ability to permeate the printing material in different directions. The anisotropic permeability coefficient is larger in the main fiber orientation direction and smaller in the perpendicular direction, and its numerical distribution is closely related to the fiber orientation vector field.

[0047] When determining the trajectory coverage of each sub-pattern unit, the boundary contour shape of the sub-pattern unit is analyzed first. For sub-pattern units with regular shapes, the trajectory coverage can be directly set to the area inside its boundary contour. For sub-pattern units with irregular shapes or depressions, convex hull processing or morphological dilation operations are required for the boundary contour to ensure that the trajectory coverage can completely cover all pixels of the sub-pattern unit. The selection of the trajectory starting point has a significant impact on the printing quality. Usually, the point with the smallest angle between the fiber orientation vector and the boundary tangent direction on the boundary contour of the sub-pattern unit is selected as the starting point. This allows the printing device to enter the sub-pattern unit along the fiber orientation direction, reducing disturbance to the fiber structure.

[0048] The calculation of trajectory density distribution requires comprehensive consideration of the spatial gradient of the fiber orientation vector field and the anisotropic penetration coefficient. The spatial gradient of the fiber orientation vector field reflects the degree of drastic change in fiber orientation in space; regions with a larger gradient indicate rapid changes in fiber orientation, requiring denser trajectories to accommodate these changes. In regions with a larger anisotropic penetration coefficient, the printing material diffuses more easily along the fiber direction, allowing for a reduction in trajectory density; conversely, in regions with a smaller anisotropic penetration coefficient, the diffusion of the printing material is limited, necessitating an increase in trajectory density to ensure uniform printing results. The trajectory density distribution can be expressed as a function of spatial location, which integrates the effects of the fiber orientation vector field gradient and the anisotropic penetration coefficient.

[0049] Track spacing directly reflects track density distribution, and its value is inversely proportional to track density. Within a sub-pattern unit, the track spacing at each position is calculated based on the track density distribution. For areas with gentle changes in fiber orientation and high anisotropic penetration coefficient, the track spacing can be set larger, typically between 0.5 mm and 2 mm; for areas with drastic changes in fiber orientation or low anisotropic penetration coefficient, the track spacing needs to be reduced to 0.2 mm to 0.8 mm to ensure that the printing material can fully cover the area. The calculation of track spacing also needs to consider the application width of the printing device to avoid uncovered gaps or excessive overlap between adjacent tracks that would lead to material accumulation.

[0050] After determining the trajectory density distribution and trajectory spacing, multiple trajectory segments are generated along the fiber orientation vector direction of the corresponding region of each sub-pattern unit. The trajectory segment generation employs a streamline tracing method. Starting from the trajectory's starting point, the segment advances one step along the fiber orientation vector direction of that point. Upon reaching a new position, the fiber orientation vector at that position is queried, and the segment continues advancing along the new fiber orientation vector direction. This process iterates until the trajectory segment reaches the boundary contour of the sub-pattern unit or intersects with other trajectory segments. The step size needs to be determined based on the spatial resolution of the fiber orientation vector field and the trajectory accuracy requirements, typically set to 0.1 mm to 0.5 mm. After generating the first trajectory segment, the starting point of the next trajectory segment is determined by offsetting a certain distance in a direction perpendicular to the fiber orientation vector according to the trajectory spacing. The streamline tracing process is repeated to generate the second trajectory segment. This process continues until a sufficient number of trajectory segments are generated within the coverage area of ​​the entire sub-pattern unit.

[0051] At the boundary between adjacent sub-pattern units, there may be significant differences in fiber orientation vectors and anisotropic penetration coefficients. When extracting fiber orientation vectors from adjacent regions at the boundary, it is necessary to obtain fiber orientation vectors within a certain range on both sides of the boundary, typically set to 5% to 10% of the boundary length. The angular difference between the fiber orientation vectors on both sides of the boundary is calculated; this angular difference reflects the degree of discontinuity in fiber orientation at the boundary. Simultaneously, the anisotropic penetration coefficients on both sides of the boundary are extracted, and the ratio of the anisotropic penetration coefficients on both sides is calculated; this ratio reflects the difference in the penetration ability of the printing material on both sides of the boundary.

[0052] The generation of the trajectory transition segment needs to be designed based on the angle difference at the boundary and the ratio of anisotropic penetration coefficients. When the angle difference is small and the ratio of anisotropic penetration coefficients is close to 1, the trajectory transition segment can use a simple straight-line connection method, directly connecting the endpoints of the trajectory segments of adjacent sub-pattern units. When the angle difference is large, the trajectory transition segment needs to use a smooth curve connection method. Commonly used curve types include Bézier curves or spline curves. The control points of the Bézier curve are determined based on the fiber orientation vectors on both sides of the boundary, so that the tangent direction of the curve at the start and end points is consistent with the direction of the fiber orientation vectors on both sides, thereby achieving a smooth transition. When the ratio of anisotropic penetration coefficients deviates significantly from 1, the length of the trajectory transition segment needs to be adjusted. The length of the transition segment should be appropriately extended on the side with a smaller anisotropic penetration coefficient to compensate for the insufficient diffusion capacity of the printing material on that side.

[0053] The generation of the trajectory transition segment also needs to consider the motion constraints of the printing device. When the printing device executes the trajectory transition segment, its direction and speed need to change continuously to avoid abrupt changes in direction or speed. Therefore, the curvature of the trajectory transition segment needs to be controlled within the maximum radius of curvature of the printing device, typically with a minimum radius of curvature of no less than 5 mm. For trajectory transition segments with excessive curvature, the length of the transition segment needs to be increased or the control point positions adjusted to reduce the curvature value. Simultaneously, the connection between the trajectory transition segment and adjacent trajectory segments needs to ensure the continuity of the tangent direction, avoiding sharp angles or broken lines. This can be achieved by adjusting the control points of the Bézier curve or spline curve.

[0054] When connecting trajectory transition segments to trajectory segments of adjacent sub-pattern units, the location of the connection point needs to be determined. The connection point is typically chosen where the trajectory segment reaches the boundary contour of the sub-pattern unit; this location can be calculated from the intersection of the trajectory segment and the boundary contour. For each sub-pattern unit, its multiple trajectory segments may have multiple connection points at the boundary. Appropriate connection points need to be selected based on the start and end points of the trajectory transition segment. During the connection process, the start point of the trajectory transition segment coincides with the end point of the trajectory segment of the preceding sub-pattern unit, and the end point of the trajectory transition segment coincides with the start point of the trajectory segment of the following sub-pattern unit, thus forming a continuous trajectory path.

[0055] After generating the trajectory segments within all sub-pattern units and the transition segments between adjacent sub-pattern units, these trajectory segments are arranged and connected in spatial order to form a complete printing application trajectory. A complete printing application trajectory is one or more continuous spatial curves covering the entire area of ​​the target printed pattern. For a printed pattern containing multiple independent sub-pattern units, multiple independent printing application trajectories may be generated, each corresponding to one or more adjacent sub-pattern units. In actual printing operations, the printing device applies the trajectories sequentially according to these trajectories to complete the creation of the entire printed pattern.

[0056] During the generation of the printing trajectory, it is necessary to optimize the trajectory to improve printing efficiency and quality. Trajectory optimization includes trajectory sequence optimization and trajectory smoothing. The goal of trajectory sequence optimization is to reduce the idle travel distance between different trajectory segments of the printing device. This is typically achieved by using the nearest neighbor algorithm or genetic algorithm to solve the traveling salesman problem and determine the optimal trajectory execution order. The goal of trajectory smoothing is to eliminate minor fluctuations and noise in the trajectory. This is achieved by using moving average filtering or Gaussian filtering to smooth the trajectory coordinates while maintaining the overall shape and orientation characteristics of the trajectory.

[0057] In one optional embodiment, generating multiple trajectory segments along the fiber orientation vector direction of the corresponding region of each sub-pattern unit, based on the trajectory density distribution and trajectory spacing, includes: Based on the trajectory density distribution and trajectory spacing, multiple trajectory starting points are set within each sub-pattern unit; The corresponding region of each sub-pattern unit is divided into multiple grid units, and the fiber orientation vector and anisotropic permeability coefficient of each grid unit are extracted. Starting from the starting point of each trajectory, read the fiber orientation vector and anisotropic permeability coefficient of the current grid cell, determine the direction of movement based on the fiber orientation vector, determine the step length based on the anisotropic permeability coefficient, and move the step length along the direction of movement to reach the next grid cell. Establish a local coordinate system at each grid cell, take the fiber orientation vector as the principal axis direction of the local coordinate system, calculate the rotation angle between the local coordinate systems of adjacent grid cells, and sum the rotation angles to obtain the cumulative deflection of the trajectory path. When the cumulative deflection exceeds the preset deflection threshold, multiple candidate grid cells are searched around the current grid cell. The similarity between the fiber orientation vector of each candidate grid cell and the fiber orientation vector of the starting grid cell is calculated, and the candidate grid cell with the highest similarity is selected as the regression target point. Move from the current grid cell toward the target point of regression, clear the accumulated deflection to zero, and then continue moving along the fiber orientation vector direction; The coordinates of each grid cell passed during the forward movement are recorded to form a position sequence, generating a trajectory segment.

[0058] In one specific implementation, after obtaining the spatial distribution positions of each sub-pattern unit and the fiber orientation vector of the corresponding region, it is necessary to generate specific printing application trajectory segments within each sub-pattern unit. The generation process of the trajectory segments needs to comprehensively consider the trajectory density distribution, trajectory spacing, and spatial variation characteristics of the fiber orientation vector to ensure that the generated trajectory segments can maintain good directional consistency with the fiber structure of the fabric.

[0059] Multiple trajectory starting points are set within each sub-pattern unit based on trajectory density distribution and trajectory spacing. The trajectory density distribution reflects the color intensity or texture complexity of the printed pattern in different areas; areas with higher density require denser trajectory coverage to achieve a finer pattern effect. The trajectory spacing is typically set between 0.3 mm and 1.2 mm, with the specific value determined based on the fiber diameter of the flannel fabric and the target printing resolution. In the boundary region of the sub-pattern unit, multiple candidate starting point positions are evenly distributed along the boundary line according to the trajectory spacing. For the internal region, a Poisson disk sampling method based on the density field is used to generate the starting point distribution, ensuring that the minimum distance between each starting point is not less than the set trajectory spacing, while automatically increasing the number of starting points in areas with higher density. Each trajectory starting point records its position coordinates in the fabric coordinate system and is labeled with its corresponding sub-pattern unit number.

[0060] The corresponding area of ​​each sub-pattern unit is divided into multiple grid units, and the fiber orientation vector and anisotropic penetration coefficient of each grid unit are extracted. The size of the grid unit is set to a square area of ​​0.5 mm by 0.5 mm, which can capture the local variation characteristics of fiber orientation while ensuring computational efficiency. For each grid unit, the fiber orientation vector of all sampling points in the area is extracted from the fiber orientation distribution data, and the average direction of these vectors is calculated as the representative fiber orientation vector of the grid unit. At the same time, the anisotropic penetration coefficient of the grid unit is calculated based on the fiber density and the degree of order of fiber arrangement within the grid unit. The anisotropic penetration coefficient reflects the ease of penetration of the printing material in different directions at that location; the larger the value, the smaller the penetration resistance along that direction. The extracted fiber orientation vector and anisotropic penetration coefficient are stored in the attribute data structure of the grid unit for subsequent trajectory generation.

[0061] Starting from the origin of each trajectory, the fiber orientation vector and anisotropic penetration coefficient of the current grid cell are read. The direction of movement is determined based on the fiber orientation vector, and the step length is determined based on the anisotropic penetration coefficient. The direction of the fiber orientation vector is the direction of movement of the trajectory segment at that position. This ensures that the printed trajectory is consistent with the fiber arrangement direction, reducing lateral diffusion and uneven penetration of the printed material. The determination of the step length needs to consider the influence of the anisotropic penetration coefficient. When the anisotropic penetration coefficient is large, it indicates that the fiber arrangement at that position is relatively loose or the fiber orientation is highly consistent. In this case, a larger step length can be used to improve trajectory generation efficiency. (Step length) With anisotropic permeability coefficient The relationship between them can be represented as ,in The reference step size is typically set to 0.2 mm. This is the step size adjustment factor, with a value ranging from 0.1 to 0.3. Based on the calculated forward direction and step size, the grid cell moves a corresponding distance from its center along the forward direction to reach the next grid cell.

[0062] A local coordinate system is established at each grid cell, with the fiber orientation vector as the principal axis of the local coordinate system. The rotation angle between the local coordinate systems of adjacent grid cells is calculated. Establishing a local coordinate system allows for a more accurate description of the spatial variation characteristics of fiber orientation. The origin is taken as the center point of the current grid cell, and the direction of the fiber orientation vector of that grid cell is defined as the principal axis of the local coordinate system. The positive direction of the axis, perpendicular to the fiber orientation vector, is defined as follows: The positive direction of the axis. When the trajectory moves from the current grid cell to the next grid cell, the next grid cell also establishes its local coordinate system. There is a rotation angle difference between the local coordinate systems of two adjacent grid cells, and this rotation angle... This can be obtained by calculating the angle between the two fiber orientation vectors. The cumulative deflection of the trajectory path is obtained by summing the rotation angles between adjacent mesh cells along the path. The cumulative deflection reflects the overall deviation of the trajectory path from the fiber orientation direction at the starting point.

[0063] When the cumulative deflection exceeds a preset deflection threshold, multiple candidate grid cells are searched around the current grid cell. The similarity between the fiber orientation vector of each candidate grid cell and the fiber orientation vector of the starting grid cell is calculated, and the candidate grid cell with the highest similarity is selected as the regression target point. The preset deflection threshold is usually set between 15 and 30 degrees. When the cumulative deflection exceeds this threshold, it indicates that the trajectory path has deviated significantly from its intended direction, requiring orientation correction to avoid excessive curvature or deviation from the effective area of ​​the sub-pattern cell. A search is conducted in the 8 adjacent grid cells and the next 16 adjacent grid cells around the current grid cell, resulting in a total of 24 candidate grid cells. For each candidate grid cell, its fiber orientation vector is extracted, and the cosine of the angle between this vector and the fiber orientation vector of the starting grid cell is calculated as a similarity index. The calculation formula is: ,in The similarity index is the angle between the fiber orientation vector of the candidate grid cell and the fiber orientation vector of the starting grid cell. The closer the similarity index is to 1, the closer the directions of the two fiber orientation vectors are. The candidate grid cell with the highest similarity index is selected as the regression target point, which can guide the trajectory path to revert to a direction that is more consistent with the starting fiber orientation direction.

[0064] The trajectory moves from the current grid cell towards the target point, resetting the accumulated deflection to zero before continuing along the fiber orientation vector direction. The path from the current grid cell to the target point is generated using linear interpolation, discretizing the path into multiple intermediate points, each corresponding to a grid cell. Upon reaching the target point, the accumulated deflection is reset to zero, indicating that the trajectory has undergone one orientation correction. Starting from the target point, the fiber orientation vector and anisotropic permeability coefficient of that grid cell are reread, and a new direction and step length are determined using the aforementioned method, continuing to generate subsequent trajectory paths. This regression correction mechanism effectively prevents excessive deviation of the trajectory path in areas with significant fiber orientation variations, ensuring that the trajectory segment always extends along the dominant fiber orientation direction.

[0065] The coordinates of each grid cell traversed during the trajectory's movement are recorded to form a position sequence, generating a trajectory segment. Throughout the trajectory generation process, the center coordinates of each grid cell traversed by the trajectory path are recorded chronologically, forming an ordered sequence of position coordinates. This position sequence fully describes the spatial path of the trajectory segment from its starting point to its ending point. The termination conditions for a trajectory segment include the following: the trajectory path reaches the boundary region of a sub-pattern unit; the trajectory path enters an area already covered by other trajectory segments; the length of the trajectory path exceeds a preset maximum trajectory length; or the trajectory path enters an area lacking fiber orientation vector data. When any termination condition is met, the generation of the current trajectory segment stops, and the recorded position sequence is stored as the geometric description data of that trajectory segment. For each trajectory starting point, the above trajectory generation process is repeated until all trajectory starting points have completed trajectory segment generation, or all areas within the sub-pattern unit are covered by trajectory segments. The generated multiple trajectory segments together constitute the complete printing application trajectory of the sub-pattern unit, providing precise path guidance for subsequent printing application operations.

[0066] In practical applications, for areas where the fiber orientation of flannel fabric changes drastically, the trajectory segment generation process may frequently trigger regression correction mechanisms. To avoid excessive backtracking and discontinuities in the trajectory path, a smoothing step can be introduced into the trajectory generation algorithm. After the initial trajectory segment generation is completed, the position sequence is locally smoothed using a moving average or Gaussian filter to weighted average the coordinates of adjacent points, eliminating minor jitter caused by grid discretization and regression correction. The smoothing window size is typically set to 5 to 9 adjacent points, and the smoothing intensity is adaptively adjusted according to the severity of the fiber orientation change. The smoothed trajectory segment maintains good consistency with the fiber orientation and exhibits a smoother and more continuous geometric shape, which is beneficial for improving the stability of the printing process and the uniformity of printing quality.

[0067] In one optional embodiment, based on the anisotropic penetration coefficient and fiber orientation vector at corresponding positions of each trajectory segment in the printing application trajectory, the applied pressure value and applied rate value of each trajectory segment are determined, and a control parameter sequence is generated, including: Extract the anisotropic penetration coefficients at corresponding positions of each trajectory segment in the printing application trajectory, construct a penetration coefficient distribution curve, calculate the curvature value of the penetration coefficient distribution curve, and identify trajectory segments whose penetration coefficient changes exceed a preset curvature threshold as target control trajectory segments. The anisotropic permeability coefficient of the target control trajectory segment is divided into intervals, and the anisotropic permeability coefficient of each interval is mapped to the corresponding applied pressure adjustment factor. The applied pressure adjustment factor is weighted and calculated with the preset benchmark pressure value to obtain the applied pressure value of each trajectory segment. Extract the fiber orientation vectors at corresponding positions of adjacent trajectory segments in the printing application trajectory, calculate the angle between adjacent fiber orientation vectors to obtain the fiber turning angle sequence, and perform differential operation on the fiber turning angle sequence to obtain the fiber turning angular velocity; The applied rate compensation amount is determined based on the fiber turning angular velocity, and the applied rate compensation amount is superimposed on the preset reference rate value to obtain the applied rate value for each trajectory segment. The applied pressure value and application rate value are associated and bound with the position coordinates of the corresponding trajectory segment, and a sequence of control parameters is generated according to the trajectory segment order of the printing application trajectory.

[0068] In one specific implementation, after obtaining the printing application trajectory, it is necessary to determine the corresponding application pressure and application rate values ​​based on the differences in fabric characteristics at each position along the trajectory, thereby generating a sequence of control parameters that can precisely control the operation of the printing device. This process first extracts the anisotropic penetration coefficients corresponding to each trajectory segment from the data structure of the printing application trajectory. A trajectory segment refers to a spatially discretized unit of the printing application trajectory; each trajectory segment corresponds to a local region on the fabric surface, which has specific fiber orientation characteristics and penetration properties. The extracted anisotropic penetration coefficients are arranged in spatial order according to the trajectory segments, forming a one-dimensional penetration coefficient distribution curve. This curve reflects the changing trend of fabric penetration performance along the direction of the printing application trajectory.

[0069] Curvature analysis is performed on the permeability coefficient distribution curve to identify locations where drastic changes in permeability occur. The curvature value is calculated using the numerical differentiation method for discrete curves, for the first... Each sampling point has a curvature value. It can be calculated using the coordinate relationship of three adjacent points. Specifically, let the first... , , The permeability coefficients corresponding to the sampling points are respectively , , The length interval of the trajectory segment is The curvature value is The calculated curvature value is compared with the preset curvature threshold. When comparing, At that time, the trajectory segment corresponding to that position is determined as the target control trajectory segment. The fabric permeability varies significantly at the locations of these trajectory segments, requiring targeted adjustment of the applied pressure to ensure that the printed material achieves a uniform permeation effect in different permeability regions.

[0070] For the identified target control trajectory segment, its anisotropic permeability coefficient is divided into intervals. Based on the statistical distribution characteristics of the fabric permeability coefficient, the value range of the permeability coefficient is divided into several intervals, each corresponding to a different permeability performance level. For example, the permeability coefficient can be divided into high permeability, medium permeability, and low permeability intervals. In areas with high permeability coefficients, the gaps between fabric fibers are larger, and the printing material easily penetrates, requiring a reduction in applied pressure to avoid excessive penetration and pattern diffusion. In areas with low permeability coefficients, the fibers are tightly packed, requiring an increase in applied pressure to ensure that the printing material fully penetrates into the fiber interior.

[0071] The anisotropic permeability coefficients of each interval are mapped to corresponding applied pressure adjustment factors. This mapping relationship can be established using piecewise linear or nonlinear functions. Let the anisotropic permeability coefficient of a certain trajectory segment be... The corresponding applied pressure adjustment factor is Then the mapping relationship can be expressed as ,in This is a mapping function. In practical applications, the specific form of this mapping function can be obtained by fitting experimental data. For example, when the permeability coefficient is in the low permeability range, the adjustment factor is set to a larger value to increase the applied pressure; when the permeability coefficient is in the high permeability range, the adjustment factor is set to a smaller value to decrease the applied pressure.

[0072] After obtaining the applied pressure adjustment factor, it is weighted and calculated with a preset reference pressure value to obtain the applied pressure value for each trajectory segment. (Preset reference pressure value) It is a standard pressure value preset according to the fabric type and printing process requirements, representing the appropriate applied pressure under average penetration performance conditions. For the first Each trajectory segment has an applied pressure value. The calculation formula is: This weighted calculation enables adaptive adjustment of the applied pressure according to the fabric's permeability, ensuring that the printed material achieves consistent penetration depth and pattern quality across different permeability zones.

[0073] While determining the applied pressure value, it is also necessary to determine the applied rate value based on the change in fiber orientation. Fiber orientation vectors at corresponding positions of adjacent trajectory segments are extracted from the printing application trajectory, and the angle between adjacent fiber orientation vectors is calculated. Let the first... The trajectory segment and the first The fiber orientation vectors corresponding to the positions of each trajectory segment are respectively and Then the included angle between the two It can be calculated using the vector dot product, i.e. Arrange the included angles between all adjacent trajectory segments in sequence to form a fiber turning angle sequence. ,in Subtract one from the total number of trajectory segments.

[0074] Differential calculations are performed on the fiber turning angle sequence to obtain the fiber turning angular velocity. The fiber turning angular velocity reflects the rate of change in fiber orientation along the printing application trajectory. For the... At each position, the fiber turning angular velocity It can be calculated by the difference between adjacent angle values, that is... ,in This represents the spatial distance between adjacent trajectory segments. When the fiber turning angular velocity is large, it indicates that the fiber orientation is changing rapidly at that position. The printing device needs to quickly adjust the application direction to adapt to the change in fiber orientation. In this case, the application rate should be reduced to ensure application accuracy. When the fiber turning angular velocity is small, the fiber orientation changes gradually, and the application rate can be increased to improve printing efficiency.

[0075] The applied rate compensation amount is determined based on the fiber turning angular velocity. There is a negative correlation between the applied rate compensation amount and the fiber turning angular velocity; that is, the larger the fiber turning angular velocity, the smaller the applied rate compensation amount, thus reducing the actual applied rate. Let the first... The applied rate compensation amount for each trajectory segment is Then its calculation formula is: ,in This is the rate compensation coefficient, determined based on the dynamic response characteristics of the printing device and the printing accuracy requirements. The applied rate compensation amount is superimposed on the preset reference rate value to obtain the applied rate value for each trajectory segment. (Preset reference rate value) It is a standard application rate pre-set according to printing process requirements and production efficiency targets. For the first Each trajectory segment has an applied rate value. The calculation formula is: This compensation mechanism enables dynamic adjustment of the application rate as fiber orientation changes, maximizing production efficiency while ensuring printing accuracy.

[0076] After obtaining the applied pressure and application rate values ​​for each trajectory segment, these control parameters need to be associated and bound to the corresponding position coordinates of the trajectory segment. Each trajectory segment has unique position coordinates in space; these coordinates can be the two-dimensional coordinates of the trajectory segment's starting point or the coordinates of the trajectory segment's center point. The applied pressure values... , applied rate value With the Position coordinates of each trajectory segment Perform correlation to form a control parameter tuple. Following the sequence of the trajectory segments of the printing application path, all control parameter tuples are arranged sequentially to generate a control parameter sequence. This sequence of control parameters fully describes the applied pressure and application rate at each position during the printing operation, providing a data foundation for the precise control of the printing device.

[0077] The generation process of the control parameter sequence fully considers the spatial distribution characteristics of the anisotropic permeability coefficient and fiber orientation vector on the fabric surface. By adaptively adjusting the applied pressure and application rate, precise matching between printing parameters and local fabric properties is achieved. This parameter optimization method based on the microstructure characteristics of the fabric can effectively solve the problem of unstable printing quality caused by uneven fabric properties in traditional printing processes, and significantly improve the pattern clarity and color uniformity of waterless printing.

[0078] like Figure 2 The diagram shown illustrates the adaptive compensation control logic flowchart of the printing device.

[0079] In one optional embodiment, during the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence, and the deformation response data of the fabric is collected in real time. A deformation-fiber deviation field is constructed based on the deformation response data, and the deviation propagation path is identified. Predictive compensation for the control parameter sequence of the unexecuted trajectory segment includes: The printing device is driven to perform the application operation of each trajectory segment in sequence along the printing application trajectory according to the control parameter sequence. During the execution, the deformation response data of the fabric surface is collected in real time by the deformation sensor and the deformation response data is converted into a deformation response vector field. The deformation response vector field is spatially superimposed with the fiber orientation distribution data at the corresponding position of the executed trajectory segment to construct a deformation-fiber deviation field. The spatial distribution gradient of the deviation vector in the deformation-fiber deviation field is extracted, and the deviation propagation path is identified based on the spatial distribution gradient. Tracing back along the deviation propagation path to the unexecuted trajectory segment, the coupling coefficient between the deviation vector at each position point on the deviation propagation path and the fiber orientation vector corresponding to each position point is calculated. Based on the coupling coefficient, predictive compensation is performed on the applied pressure value and applied rate value of the unexecuted trajectory segment. The compensated applied pressure and applied rate values ​​are updated to the control parameters of the corresponding unexecuted trajectory segment in the control parameter sequence, and the printing device continues to be driven to perform the application operation of the remaining trajectory segment according to the updated control parameter sequence.

[0080] In one specific embodiment, during the execution phase of the waterless printing operation, the printing device applies the printing material along the printing application trajectory according to a pre-generated sequence of control parameters. The control parameter sequence includes the application pressure and application rate values ​​for each trajectory segment, which have been preliminarily calculated based on the fiber orientation vector and anisotropic penetration coefficient before execution. The printing device is equipped with a deformation sensor array arranged at multiple monitoring positions on the fabric surface, capable of capturing the deformation response of the fabric during the application process in real time. The deformation sensors employ optical displacement sensors or capacitive displacement sensors, achieving measurement accuracy at the micrometer level, with sampling frequencies set to 100Hz to 500Hz to ensure the capture of transient deformation characteristics of the fabric surface during application.

[0081] When the printing device performs the first During the application of an operation on a trajectory segment, the deformation sensor array synchronously acquires deformation response data of the surrounding area. The deformation response data includes the displacement and direction of each monitoring point on the fabric surface. These discrete deformation response data are then converted into a continuous deformation response vector field. The conversion process employs radial basis function interpolation for any position on the fabric surface. Its deformation response vector The deformation response vector is obtained by weighted interpolation of deformation data from multiple surrounding monitoring points. The magnitude of the deformation response vector reflects the degree of deformation at that location, while the direction reflects the dominant direction of deformation.

[0082] After the deformation response vector field is constructed, it is spatially superimposed with the fiber orientation distribution data at the corresponding positions of the executed trajectory segments. The fiber orientation distribution data was obtained through fiber orientation detection before the printing operation began, recording the fiber orientation vectors at various positions on the fabric surface. For the executed... For each trajectory segment, the fiber orientation vector field of the area covered by the trajectory segment and its surrounding influence area is extracted. and deformation response vector field In each spatial location At this point, calculate the deviation vector between the deformation response vector and the fiber orientation vector. The deviation vector is defined as the component of the deformation response vector perpendicular to the fiber orientation direction. The formula for calculating the deviation vector is: ,in This represents the projection of the deformation response vector onto the fiber orientation direction.

[0083] A deformation-fiber deviation field is constructed by calculating the deviation vectors at all locations on the fabric surface. This deviation field reflects the degree of mismatch between the deformation caused by the applied operation and the fiber orientation. Regions with large deviation vector amplitudes indicate a significant deviation between the deformation direction and the fiber orientation direction in that region. This deviation can lead to uneven printing results and pattern distortion. To identify the propagation characteristics of the deviation on the fabric surface, the spatial distribution gradient of the deviation vectors in the deformation-fiber deviation field is extracted. The spatial distribution gradient is obtained by calculating the deviation vector amplitudes. The spatial derivative is obtained, and the gradient direction points in the direction where the magnitude of the deviation vector increases the fastest. The gradient magnitude reflects the drastic degree of deviation change.

[0084] The deviation propagation path is identified based on the spatial distribution gradient. The deviation propagation path refers to the path along which the deviation vector diffuses and accumulates on the fabric surface in a specific direction. This path typically extends along regions where fiber orientation changes abruptly or where the permeability coefficient gradient is large. The identification process employs a gradient tracking algorithm, starting from the application position of the executed trajectory segment and tracing along the direction of the spatial distribution gradient of the deviation vector. The tracking step size is set to half the fabric grid cell size. At each tracking step position, the magnitude and gradient direction of the deviation vector at that position are calculated. When the deviation vector magnitude exceeds a preset threshold... At this point, the location is marked as a critical node on the deviation propagation path. The tracking process continues until the deviation vector amplitude drops below a threshold or the tracking path reaches the fabric boundary.

[0085] After the deviation propagation path is identified, it is traced back along the path to the unexecuted trajectory segment. The tracing process determines whether the deviation propagation path intersects with or is adjacent to the spatial location of the unexecuted trajectory segment. If the deviation propagation path passes through or approaches an unexecuted trajectory segment, the operation applied to that trajectory segment may be affected by the generated deviation. For the affected unexecuted trajectory segment, predictive compensation of its control parameters is required to offset the negative effects of deviation propagation.

[0086] Predictive compensation is calculated based on the coupling relationship between the deviation vector and the fiber orientation vector at each location point along the deviation propagation path. For the first... There are 10 location points, and their deviation vectors are: The fiber orientation vector is Define the coupling coefficient at this location point. The cosine of the angle between the deviation vector and the fiber orientation vector is calculated using the following formula: The larger the absolute value of the coupling coefficient, the stronger the consistency between the deviation vector and the fiber orientation direction, and the higher the efficiency of deviation propagation along the fiber direction. When the coupling coefficient is positive, the deviation vector is in the same direction as the fiber orientation direction, indicating that the deformation is stretching along the fiber direction; when the coupling coefficient is negative, the deviation vector is in the opposite direction to the fiber orientation direction, indicating that the deformation is compressing along the fiber direction.

[0087] For unexecuted trajectory segments affected by deviation propagation, the applied pressure compensation and applied rate compensation are calculated based on the coupling coefficients of each location point along the deviation propagation path. (Applied pressure compensation amount) The calculation considers the weighted average of the coupling coefficients at all points along the deviation propagation path, and the calculation formula is as follows: ,in This represents the total number of locations along the deviation propagation path. For the first The weight coefficient of each location point is inversely proportional to the distance from that location point to the unexecuted trajectory segment. This is the pressure compensation gain coefficient, ranging from 0.1 to 0.5. When the deviation vector is stretched along the fiber direction, the applied pressure compensation is positive, indicating that the applied pressure needs to be increased to overcome the tensile resistance of the fabric; when the deviation vector is compressed along the fiber direction, the applied pressure compensation is negative, indicating that the applied pressure needs to be reduced to avoid over-compression.

[0088] Apply rate compensation amount The calculation is based on the spatial gradient characteristics of the deviation propagation path. A larger spatial gradient magnitude indicates a more drastic change in deviation, requiring a corresponding adjustment of the application rate to adapt to the fabric's dynamic response characteristics. The formula for calculating the application rate compensation is as follows: ,in For the first The gradient magnitude of the deviation vector at each location point This is the rate compensation gain coefficient, ranging from 0.05 to 0.2. Applying a negative rate compensation value indicates that the application rate needs to be reduced in regions with large deviation gradients to slow down the propagation of deformation and improve the stability of the application process.

[0089] The calculated applied pressure compensation and applied rate compensation are superimposed on the original control parameters of the unexecuted trajectory segment to obtain the compensated applied pressure value. and applied rate value The compensated control parameters are updated to the corresponding unexecuted trajectory segments in the control parameter sequence, replacing the original control parameters. The update process adopts a real-time update strategy. After each trajectory segment is applied, the control parameters for subsequent unexecuted trajectory segments are immediately recalculated and updated to ensure that the control parameter sequence always reflects the latest fabric deformation state and deviation propagation characteristics.

[0090] After the control parameter sequence is updated, the printing device continues to apply the remaining trajectory segment according to the updated control parameter sequence. During execution, the deformation sensor continuously collects the fabric's deformation response data, repeating the aforementioned processes of constructing the deformation response vector field, constructing the deformation-fiber deviation field, identifying the deviation propagation path, and predictive compensation, forming a closed-loop feedback control mechanism. Through real-time monitoring and dynamic compensation, the accumulation and propagation of deformation deviations during application are effectively suppressed, ensuring the uniformity and consistency of the printed pattern across the entire fabric surface, and improving the processing quality and pattern fidelity of waterless printing.

[0091] A second aspect of the present invention provides an intelligent control system for waterless printing of flannel, comprising: The data acquisition unit is used to acquire fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; The pattern decomposition unit is used to calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data. Based on the fiber orientation vector and penetration coefficient, the target printed pattern data is decomposed in an adaptive direction to obtain multiple sub-pattern units corresponding to different fiber orientation regions. The trajectory generation unit is used to generate a printing application trajectory based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area; The parameter determination unit is used to determine the application pressure and application rate values ​​of each trajectory segment based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, and to generate a sequence of control parameters. The compensation control unit is used to drive the printing device to apply the printing along the printing application trajectory according to the control parameter sequence during the waterless printing operation, collect the deformation response data of the fabric in real time, construct the deformation-fiber deviation field based on the deformation response data and identify the deviation propagation path, and perform predictive compensation for the control parameter sequence of the unexecuted trajectory segment.

[0092] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0093] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0094] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart control method for waterless printing of flannel, characterized in that, include: Obtain fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; Calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data, and perform directional adaptive decomposition on the target printed pattern data based on the fiber orientation vector and penetration coefficient to obtain multiple sub-pattern units corresponding to different fiber orientation regions. Based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area, a printing application trajectory is generated; Based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, the applied pressure value and applied rate value of each trajectory segment are determined, and a control parameter sequence is generated. In the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence. The deformation response data of the fabric is collected in real time. Based on the deformation response data, a deformation-fiber deviation field is constructed and the deviation propagation path is identified. Predictive compensation is performed on the control parameter sequence of the unexecuted trajectory segment.

2. The method according to claim 1, characterized in that, Based on the fiber orientation distribution data, the fiber orientation vector and penetration coefficient of each region on the fabric surface are calculated. Then, the target printed pattern data is subjected to directional adaptive decomposition based on the fiber orientation vector and penetration coefficient, resulting in multiple sub-pattern units corresponding to different fiber orientation regions, including: Extract fiber orientation angle and fiber density data from each sampling point in the fiber orientation distribution data. Construct a fiber orientation vector field for each region based on the fiber orientation angle. Calculate the anisotropic permeability coefficient and directional permeability resistance coefficient for each region based on the fiber density data, the spatial gradient of the fiber orientation vector field, and the angle between fiber orientation vectors of adjacent regions. Extract the pattern edge direction distribution and the internal texture direction distribution of the target printed pattern data, and construct the direction feature tensor field; Spatially register the orientation feature tensor field with the fiber orientation vector field, calculate the orientation fit index at each spatial location, and determine the decomposition boundary position and decomposition hierarchy structure of the target printed pattern data based on the spatial distribution characteristics of the orientation fit index. The target printed pattern data is decomposed along the decomposition boundary to obtain multiple sub-pattern units. The pattern orientation features of each sub-pattern unit are coordinated with the fiber orientation vector and permeability coefficient distribution characteristics of the corresponding fiber orientation region. Furthermore, at the decomposition boundary, the pattern orientation features of adjacent sub-pattern units match the fiber orientation vector conversion angle at the corresponding position.

3. The method according to claim 2, characterized in that, Spatially register the orientation feature tensor field with the fiber orientation vector field, calculate the orientation fit index at each spatial location, and determine the decomposition boundary position and decomposition hierarchy structure of the target printed pattern data based on the spatial distribution characteristics of the orientation fit index, including: Spatial alignment and registration are performed on the orientation feature tensor field and the fiber orientation vector field; Extract the principal direction components of the orientation feature tensor field at each spatial location, calculate the angle deviation between the principal direction components and the corresponding fiber orientation vector, and calculate the orientation coupling strength at each spatial location based on the orientation dispersion of the orientation feature tensor field at each spatial location and the spatial gradient of the fiber orientation vector field. Based on the included angle deviation and the directional coupling strength, and combined with the anisotropic permeability coefficient at the corresponding position, the directional fit index of each spatial position is calculated. Identify the spatial gradient abrupt change locations of the orientation fitness index and the connected regions where the orientation fitness index is lower than a preset fitness threshold. Determine the spatial gradient abrupt change locations as first-level decomposition boundaries and the boundaries of the connected regions as second-level decomposition boundaries. Based on the spatial topological relationship between the first-level decomposition boundary and the second-level decomposition boundary, a decomposition hierarchy structure containing a first decomposition level and a second decomposition level is constructed, wherein the first decomposition level corresponds to the region divided by the first-level decomposition boundary, and the second decomposition level corresponds to the refined division region of the second-level decomposition boundary within the first decomposition level.

4. The method according to claim 1, characterized in that, Based on the spatial distribution of multiple sub-pattern units and the fiber orientation vectors of the corresponding regions, the printing application trajectory is generated, including: The spatial distribution and boundary contours of multiple sub-pattern units are obtained, and the fiber orientation vector field and anisotropic permeability coefficient distribution of the corresponding region of each sub-pattern unit are extracted. Based on the boundary contour of each sub-pattern unit and the fiber orientation vector field of the corresponding region, the trajectory coverage and trajectory starting point of each sub-pattern unit are determined, and the trajectory density distribution and trajectory spacing inside each sub-pattern unit are calculated based on the spatial gradient and anisotropic penetration coefficient of the fiber orientation vector field. Based on the trajectory density distribution and trajectory spacing, multiple trajectory segments are generated along the fiber orientation vector direction of the corresponding region of each sub-pattern unit; Extract the fiber orientation vector and anisotropic penetration coefficient at the boundary of adjacent sub-pattern units. Generate a trajectory transition segment based on the angular difference between the fiber orientation vectors of adjacent regions at the boundary and the anisotropic penetration coefficient. Connect the trajectory transition segment to the trajectory segment of adjacent sub-pattern units to form a complete printing application trajectory.

5. The method according to claim 4, characterized in that, Based on the trajectory density distribution and trajectory spacing, multiple trajectory segments are generated along the fiber orientation vector direction of the corresponding region of each sub-pattern unit, including: Based on the trajectory density distribution and trajectory spacing, multiple trajectory starting points are set within each sub-pattern unit; The corresponding region of each sub-pattern unit is divided into multiple grid units, and the fiber orientation vector and anisotropic permeability coefficient of each grid unit are extracted. Starting from the starting point of each trajectory, read the fiber orientation vector and anisotropic permeability coefficient of the current grid cell, determine the direction of movement based on the fiber orientation vector, determine the step length based on the anisotropic permeability coefficient, and move the step length along the direction of movement to reach the next grid cell. Establish a local coordinate system at each grid cell, take the fiber orientation vector as the principal axis direction of the local coordinate system, calculate the rotation angle between the local coordinate systems of adjacent grid cells, and sum the rotation angles to obtain the cumulative deflection of the trajectory path. When the cumulative deflection exceeds the preset deflection threshold, multiple candidate grid cells are searched around the current grid cell. The similarity between the fiber orientation vector of each candidate grid cell and the fiber orientation vector of the starting grid cell is calculated, and the candidate grid cell with the highest similarity is selected as the regression target point. Move from the current grid cell toward the target point of regression, clear the accumulated deflection to zero, and then continue moving along the fiber orientation vector direction; The coordinates of each grid cell passed during the forward movement are recorded to form a position sequence, generating a trajectory segment.

6. The method according to claim 1, characterized in that, Based on the anisotropic penetration coefficient and fiber orientation vector at corresponding positions in each trajectory segment of the printing application trajectory, the application pressure and application rate values ​​for each trajectory segment are determined, generating a sequence of control parameters including: Extract the anisotropic penetration coefficients at corresponding positions of each trajectory segment in the printing application trajectory, construct a penetration coefficient distribution curve, calculate the curvature value of the penetration coefficient distribution curve, and identify trajectory segments whose penetration coefficient changes exceed a preset curvature threshold as target control trajectory segments. The anisotropic permeability coefficient of the target control trajectory segment is divided into intervals, and the anisotropic permeability coefficient of each interval is mapped to the corresponding applied pressure adjustment factor. The applied pressure adjustment factor is weighted and calculated with the preset benchmark pressure value to obtain the applied pressure value of each trajectory segment. Extract the fiber orientation vectors at corresponding positions of adjacent trajectory segments in the printing application trajectory, calculate the angle between adjacent fiber orientation vectors to obtain the fiber turning angle sequence, and perform differential operation on the fiber turning angle sequence to obtain the fiber turning angular velocity; The applied rate compensation amount is determined based on the fiber turning angular velocity, and the applied rate compensation amount is superimposed on the preset reference rate value to obtain the applied rate value for each trajectory segment. The applied pressure value and application rate value are associated and bound with the position coordinates of the corresponding trajectory segment, and a sequence of control parameters is generated according to the trajectory segment order of the printing application trajectory.

7. The method according to claim 1, characterized in that, In the waterless printing operation, the printing device is driven to apply the printing along the printing application trajectory according to the control parameter sequence. Deformation response data of the fabric is collected in real time. Based on the deformation response data, a deformation-fiber deviation field is constructed and the deviation propagation path is identified. Predictive compensation is performed on the control parameter sequence for the unexecuted trajectory segment, including: The printing device is driven to perform the application operation of each trajectory segment in sequence along the printing application trajectory according to the control parameter sequence. During the execution, the deformation response data of the fabric surface is collected in real time by the deformation sensor and the deformation response data is converted into a deformation response vector field. The deformation response vector field is spatially superimposed with the fiber orientation distribution data at the corresponding position of the executed trajectory segment to construct a deformation-fiber deviation field. The spatial distribution gradient of the deviation vector in the deformation-fiber deviation field is extracted, and the deviation propagation path is identified based on the spatial distribution gradient. Tracing back along the deviation propagation path to the unexecuted trajectory segment, the coupling coefficient between the deviation vector at each position point on the deviation propagation path and the fiber orientation vector corresponding to each position point is calculated. Based on the coupling coefficient, predictive compensation is performed on the applied pressure value and applied rate value of the unexecuted trajectory segment. The compensated applied pressure and applied rate values ​​are updated to the control parameters of the corresponding unexecuted trajectory segment in the control parameter sequence, and the printing device continues to be driven to perform the application operation of the remaining trajectory segment according to the updated control parameter sequence.

8. A smart control system for waterless printing of flannel, used to implement the method as described in any one of claims 1-7, characterized in that, include: The data acquisition unit is used to acquire fiber orientation distribution data and target printing pattern data of the flannel fabric to be printed; The pattern decomposition unit is used to calculate the fiber orientation vector and penetration coefficient of each region on the fabric surface based on the fiber orientation distribution data. Based on the fiber orientation vector and penetration coefficient, the target printed pattern data is decomposed in an adaptive direction to obtain multiple sub-pattern units corresponding to different fiber orientation regions. The trajectory generation unit is used to generate a printing application trajectory based on the spatial distribution of multiple sub-pattern units and the fiber orientation vector of the corresponding area; The parameter determination unit is used to determine the application pressure and application rate values ​​of each trajectory segment based on the anisotropic penetration coefficient and fiber orientation vector at the corresponding positions of each trajectory segment in the printing application trajectory, and to generate a sequence of control parameters. The compensation control unit is used to drive the printing device to apply the printing along the printing application trajectory according to the control parameter sequence during the waterless printing operation, collect the deformation response data of the fabric in real time, construct the deformation-fiber deviation field based on the deformation response data and identify the deviation propagation path, and perform predictive compensation for the control parameter sequence of the unexecuted trajectory segment.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images