Method and device for adaptive adjustment of printing parameters of a laser printer

Abstract

Embodiments of the present application disclose a method and device for self-adaptive adjustment of printing parameters of a laser printer. The method can include: obtaining preference information of a printing user of a target file; the preference information is used to determine at least a first scale of printing parameters; using an artificial intelligence (AI) model to identify a first feature of the target file; the first feature includes a pixel level feature and / or a sub-pixel level feature; the first feature is used to determine at least a second scale of printing parameters; the second scale is smaller than the first scale; and determining printing parameters for printing the target file according to at least two of the preference information, currently supported printing parameters of the laser printer and the first feature of the target file.

Description

Technical Field

[0001] This invention relates to the field of laser technology, and in particular to a method and apparatus for adaptive adjustment of printing parameters of a laser printer. Background Technology

[0002] In current laser printing technologies, printing parameters are primarily set using traditional settings. For example, users pre-set preferences such as paper size, print quality, and color mode, and the system directly applies these parameters for printing. However, different users have different printing preferences for the same type of file (such as documents and images), and even the same user may have different printing needs for different types of files (such as PowerPoint presentations and Word documents). Existing technologies cannot dynamically identify file types or content characteristics, resulting in print quality that fails to meet personalized needs. Furthermore, these settings are not intelligent enough and largely rely on manual user operation. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for adaptive adjustment of printing parameters of a laser printer. The technical solution of the present invention is implemented as follows:

[0004] A first aspect of this disclosure provides a method for adaptively adjusting printing parameters of a laser printer, characterized in that the method includes: determining the printing user's preference information for a target document; the preference information being used at least to determine printing parameters at a first scale; using an artificial intelligence (AI) model to identify a first feature of the target document; the first feature including pixel-level features and / or sub-pixel-level features; the first feature being used at least to determine printing parameters at a second scale; the second scale being smaller than the first scale; and determining the printing parameters for printing the target document based on at least two of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target document.

[0005] A second aspect of this disclosure provides an adaptive adjustment device for printing parameters of a laser printer, comprising: a first determining module, configured to determine the user's preference information for printing a target document; the preference information being used to determine printing parameters at least at a first scale; an identification module, configured to identify a first feature of the target document using an artificial intelligence (AI) model; the first feature including pixel-level features and / or sub-pixel-level features; the first feature being used to determine printing parameters at least at a second scale; the second scale being smaller than the first scale; and a second determining module, configured to determine printing parameters for printing the target document based on at least two of the preference information, printing parameters currently supported by the laser printer, and the first feature of the target document.

[0006] By combining user preference information, file content characteristics, and the current printer status, the embodiments of the present invention achieve the following technical effects:

[0007] Multi-dimensional intelligent optimization of printing parameters achieves fine-tuning by integrating users' historical printing preferences with document content features (pixel-level / subpixel-level) identified by AI models. For example, high-resolution printing is automatically applied to areas containing high-precision images, while standard resolution is used for plain text areas, significantly improving printing efficiency while ensuring print quality.

[0008] The enhanced personalized printing experience dynamically adjusts printing parameters based on user preferences, meeting the diverse needs of different users for the same document. For example, designers may prefer high-fidelity color reproduction, while office users prioritize printing speed; the system can automatically adapt to the preset preferences of different users.

[0009] This enhances device compatibility and extends lifespan. It dynamically adjusts printing strategies based on currently supported printer parameters (such as toner level and printhead status) to prevent print quality degradation or device wear and tear due to parameter mismatches. For example, when insufficient toner is detected, it automatically reduces color saturation in non-critical areas.

[0010] AI-driven content awareness capabilities utilize deep learning models to perform pixel-level analysis of document content, identifying different elements such as text, images, and charts, and optimizing printing parameters accordingly. For example, it automatically enhances the contrast of small-sized text and sharpens blurry image areas. Attached Figure Description

[0011] Figure 1 A flowchart illustrating an adaptive adjustment method for printing parameters of a laser printer provided in an embodiment of the present invention;

[0012] Figure 2 A flowchart illustrating an adaptive adjustment method for printing parameters of a laser printer provided in an embodiment of the present invention;

[0013] Figure 3 A schematic diagram of the structure of an adaptive adjustment device for printing parameters of a laser printer provided in an embodiment of the present invention;

[0014] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0016] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein.

[0017] like Figure 1 As shown in the figure, this disclosure provides a method for adaptive adjustment of printing parameters of a laser printer, the method comprising:

[0018] S1110: Determine the printing user's preference information for the target file; the preference information is used to determine at least the printing parameters for the first scale;

[0019] S1120: Use an artificial intelligence (AI) model to identify a first feature of the target file; the first feature includes pixel-level features and / or sub-pixel-level features; the first feature is used to determine printing parameters at least at a second scale; the second scale is smaller than the first scale;

[0020] S1130: Determine the printing parameters for printing the target file based on at least two of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file.

[0021] In some embodiments, printing user preference information includes personalized settings reflected in the user's historical printing behavior, such as color mode (e.g., always using cool tones), resolution preference (e.g., document default 300dpi, photo default 600dpi), and paper type preference (e.g., preferring matte paper).

[0022] In some embodiments, the first scale can be a global parameter, such as DPI, color mode, or paper size.

[0023] In other embodiments, the second scale can be a local optimization parameter, such as contrast enhancement of a specific area or sharpening of small text.

[0024] In some embodiments, pixel-level features may include: edges and textures in the image, such as text strokes with a width of <0.5mm;

[0025] In other embodiments, subpixel-level features may include finer details, such as halftone dot density distribution.

[0026] For example, the DPI (dots per inch) of a laser printer directly affects printing accuracy. The solution needs to dynamically adjust output parameters based on the printer's highest supported resolution (e.g., 1200 dpi for enterprise-level devices, but only 600 dpi for home devices). For instance, if a user prefers a higher resolution (e.g., 600 dpi) but the printer only supports 300 dpi, a dithering algorithm can simulate higher accuracy. In this way, software can compensate for print quality issues caused by insufficient hardware capabilities, resulting in print quality falling short of user expectations.

[0027] In some embodiments, the AI ​​model may use a high-contrast mode for text regions, such as increasing the contrast of character edges, and leveraging the high-contrast imaging capabilities of the printer hardware to improve readability.

[0028] In some embodiments, the AI ​​model's linkage with color modes may include, but is not limited to:

[0029] If the printer supports CMYK color separation printing, the AI ​​model can identify color areas in the image and prioritize printing with primary colors (such as in photo scenes) rather than the default grayscale mode.

[0030] If a monochrome laser printer is used, it should be forced to convert color images to 16 grayscale levels to at least maintain the grayscale of different colors and reduce grayscale distortion.

[0031] In some embodiments, paper type detection and adaptation may include, but are not limited to, automatically matching printing parameters using a paper sensor built into the printer, such as an optical sensor to detect thickness and a capacitive sensor to detect humidity. For example, for thick paper (such as 200g coated paper): reduce the printing speed (from 25 pages / minute to 15 pages / minute), increase the fixing temperature (from 180°C to 200°C), and prevent paper jams.

[0032] For example, for special paper (such as self-adhesive labels): enable label mode and adjust the laser power to 50% to avoid penetrating the paper.

[0033] In some embodiments, dynamic paper feed control may include, but is not limited to: if paper skew is detected (paper feed angle deviation fed back by the sensor >2°), automatically pausing printing and prompting the user to reload the paper to avoid edge cutting errors caused by paper misalignment.

[0034] In some embodiments, the AI ​​model also optimizes the printing order, which can be used as a primary printing parameter. For example, the AI ​​model adjusts the printing queue order based on the urgency level in the user's preferences (e.g., urgent documents). Another example is enabling a fast mode for high-priority tasks, skipping preview and printing directly, and allocating a separate memory buffer to avoid conflicts with other tasks. In some embodiments, the AI ​​model determines that regular tasks should be printed in batches, using page merging technology (combining multiple pages into a single page) to improve paper utilization.

[0035] In some embodiments, the AI ​​model is configured with an error retry mechanism, which may include: if a paper jam occurs (the sensor detects that the paper has been stuck for more than 5 seconds), automatically reversing the paper and retrying to print the current page, up to 3 times. If the toner is low, the user is prompted to replace the toner cartridge, and an estimated number of pages remaining is provided and sent to the user or printer administrator. For example, the AI ​​model may calculate the estimated number of pages remaining based on the current toner concentration.

[0036] In some embodiments, the AI ​​model also coordinates feature recognition with hardware control, which may include dynamic optimization of text / image regions. For example, when the AI ​​model recognizes a table region (regular grid structure), it enables the vector printing mode of the laser printer to draw lines at 1200 dpi resolution, avoiding blurring caused by rasterization. As another example, for irregular content such as handwritten signatures, it switches to high-resolution mode and magnifies a local area (120% scaling) to utilize the printer's high-precision imaging capabilities to restore details.

[0037] In some embodiments, for printed materials with restricted rights, the AI ​​model, after identification, can embed anti-counterfeiting features to facilitate subsequent tracking and leakage tracing. For example, optical anti-counterfeiting marks (such as microtext or QR codes) can be automatically added to the printed document, achieving uncopyability through printer hardware. Exemplarily, variable frequency laser technology can be used to generate dynamic QR codes that are only visible from specific angles. This is particularly relevant for bank printing scenarios, aiming to reduce information security issues. By introducing such optical anti-counterfeiting marks, banks can perform print verification and validation.

[0038] For example, for important documents (such as contracts), a special watermark can be added to the header / footer, forming an invisible pattern through minute fluctuations (±5%) in laser power.

[0039] In some embodiments, the AI ​​model also considers energy consumption and environmental management. For example, it dynamically adjusts laser power based on print load. For instance, during single-page printing, it reduces laser power to 30% (energy-saving mode) to extend drum life. Another example is enabling fuser preheating buffer during batch printing to reduce energy consumption from repeated heating. Yet another example is automatically switching to ECOPASSIVE mode for color laser printers to reduce VOC (volatile organic compound) emissions.

[0040] In some embodiments, S1110 may include: synchronizing historical preferences from a local database or the cloud when the user prints for the first time; if there is no historical data, enabling default parameters (such as standard office mode).

[0041] In some embodiments, if it is detected that the user has modified the default settings (such as manually selecting high contrast), the preference information is updated in real time and S1110 is re-executed.

[0042] In some embodiments, S1120 may include: the target file is an image / mixed document (such as a PDF containing scanned images); the file resolution is ≥300dpi (low-resolution files skip subpixel analysis).

[0043] Dynamic feature weight allocation: Adjust feature importance according to file type (e.g., prioritize text edges in contract documents, and prioritize color transitions in photos).

[0044] Real-time feedback mechanism: If the confidence level of the AI ​​model is less than the threshold, manual annotation will be triggered (e.g., a pop-up window will ask if the table lines need to be enhanced).

[0045] In one embodiment, S1130 can perform a comprehensive parameter decision to obtain the printing parameters for this print or batch printing. For example, the decision matrix can be as shown in Table 1 below:

[0046] Combination of conditions Print parameter priority User preferences + AI features 70% preference + 30% AI-generated parameters AI features only The parameters are 100% AI-generated, but this example requires user confirmation. User preferences + printer restrictions Find the intersection (e.g., if the user wants 600dpi but the printer only supports 300dpi, downgrade and prompt the user).

[0047] Table 1

[0048] In some embodiments, if the printer does not support AI-generated parameters (e.g., it requires 10-micron droplets but the device only supports 20-micron droplets), it will fall back to user preferences or default parameters and log the conflict.

[0049] In some embodiments, the AI ​​model includes an adaptive learning module. This module records user feedback each time a printout occurs (e.g., too dark / too blurry) and dynamically adjusts the feature weights of the AI ​​model (e.g., reducing contrast enhancement weights for subsequent printouts of the same user).

[0050] In some embodiments, cross-scale parameter fusion may include, but is not limited to: combining global parameters (such as DPI) of the first scale with local optimizations (such as text sharpening areas) of the second scale to generate layered printing instructions (such as using 1200 dpi for the background and an additional 400% sharpening for the text area).

[0051] The following is an example using a specific application scenario. A user prints a scanned copy of a contract (including a form and a handwritten signature). In step S1110, historical data reveals that the user prefers a high-contrast + black-and-white mode for contract documents. Step S1120 is executed, where AI identifies:

[0052] Pixel-level features: table lines are 2px wide, and handwritten signature strokes are discontinuous;

[0053] Subpixel-level feature: uneven ink density at the signature area.

[0054] Implement the S1130 integrated decision: adopt the user-preferred black and white mode (first scale); enable AI-generated local contrast enhancement for the signature area (second scale). Assume that because the printer does not support subpixel-level droplet control, 300% software sharpening will be used instead.

[0055] In some embodiments, the AI ​​model will also automatically identify sensitive information. For example, if it detects that a file contains sensitive information (such as identification), it will automatically trigger an information protection mode: forcibly reduce the resolution to 150 dpi (to prevent microfilm restoration); and disable the AI's detail enhancement function (to avoid highlighting character edges).

[0056] Table 2 below illustrates the comprehensive decision-making process of another AI model:

[0057] Execute operation Triggering conditions Hardware dependency Jump path Paper type detection The sensor detects paper insertion when the print job begins. Optical / capacitive sensors Normal → Print; Abnormal → Pause and display a message High-precision mode enabled AI can identify fine details (such as lines smaller than 0.5mm). High-resolution laser head, variable aperture Switch to 1200dpi mode Automatic fault repair Three consecutive paper jams or low toner alarms Fault sensors, remote diagnostic modules Transfer to human assistance process

[0058] Table 2

[0059] In some embodiments, when the user prefers an ink-saving mode, printing specific subpixel features (such as 10-15μm dot matrix) uses reverse exposure enhancement to save on laser printing consumables (e.g., toner consumption). For pixel-level features or higher, reducing exposure intermittently and instead using reverse exposure enhancement while in ink-saving mode achieves the effect of reducing consumables. This is due to the critical effect of printing subpixel-level features. Printing subpixel-level features can be used for high-requirement target document printing such as anti-counterfeiting invoices and medical images.

[0060] In some embodiments, when the user's preferred quality level is higher than a specified level, the calculation weight of sub-pixel level features is increased when determining printing parameters to calculate more suitable printing parameters.

[0061] In one embodiment, an AI model (such as CNN / Transformer) is used to analyze the subpixel-level characteristics of a file (such as gradient distribution and edge energy), not limited to text / image ternary classification. For example, recognizing gradient images requires satisfying the Laplacian smoothness constraint. Dynamically increase the dot matrix density.

[0062] Co-mapping of preferences and features corresponds to the coordinates of the image. The laser energy at the corresponding edge can be as follows:

[0063] Establish a joint mapping function between user preferences (such as high contrast) and pixel features (such as handwritten stroke energy). ,in, For user preference matrix, Pixel edge energy.

[0064] In some embodiments, the printing parameters for the first scale include at least one of the following: page settings; color type settings; and / or, the printing parameters for the second scale include at least one of the following: printing resolution, laser beam wavelength, laser pulse repetition frequency, laser scanning speed, laser dot density, laser dot power, and number of printing laser groups.

[0065] In some embodiments, the rapid response of hardware adaptation at the first scale and the fine-grained optimization driven by AI at the second scale not only leverage the physical performance advantages of laser printers but also overcome the physical limitations of hardware through software algorithms. This achieves an optimal balance of speed, quality, and cost, especially in scenarios involving high-precision documents, complex images, and printing on special materials, demonstrating significant engineering application value. More importantly, the second-scale printing parameters involve adaptive adjustment of laser parameters, a breakthrough for ordinary users. This avoids the problem of using a single laser parameter after setting the laser parameters of a laser printer without professional assistance, leading to high wear on individual lasers and failing to fully utilize the capabilities of the laser printer.

[0066] In some embodiments, such as Figure 2 As shown, the step of using an artificial intelligence (AI) model to identify the pixel-level features of the target file includes:

[0067] S1121: The pre-recognition layer of the AI ​​model identifies the image of the target file and determines whether the image needs to be subjected to sub-pixel level feature recognition;

[0068] S1122: When the imaging requires sub-pixel level feature recognition, the imaging is input into the first branch of the AI ​​model;

[0069] S1123: The first branch extracts the sub-pixel level features of the image;

[0070] S1124: When the imaging does not require sub-pixel level feature recognition, the imaging is input into the second branch of the AI ​​model;

[0071] S1125: The second branch extracts the pixel-level features of the image.

[0072] In some embodiments, the pre-identification layer makes precise judgments: The pre-identification layer quickly analyzes the global features of the image (such as edge density and texture complexity) using a lightweight convolutional neural network (such as MobileNetV3) to determine whether sub-pixel-level analysis needs to be initiated. For example: Scenarios requiring sub-pixel analysis include: tiny handwritten signatures in contracts (stroke width < 0.3 mm) and gradient color areas in high-resolution photos. Scenarios requiring only pixel analysis include: plain text documents and low-resolution scans (below 300 dpi).

[0073] In some embodiments, for imaging that does not require subpixel analysis (such as plain text), the complex calculations of the first branch (such as Hessian matrix gradient calculation) are skipped directly, reducing GPU / CPU load by more than 30% and improving overall processing speed.

[0074] Enhancing printing accuracy through sub-pixel level feature extraction may include, but is not limited to:

[0075] Refinement of the first branch: The first branch uses a high-precision network (such as U-Net++) to extract sub-pixel level features (such as halftone dot density and 0.01mm grayscale transition of text stroke edges) and outputs a high-dimensional feature map (such as 16x super-resolution features).

[0076] Laser printer hardware collaboration: Mapping subpixel features to printer hardware capabilities (such as 600dpi→1200dpi interpolation printing, dynamic adjustment of laser pulse frequency) to achieve: Text sharpening: Subpixel-level compensation for stroke edges of small-sized text (such as 6pt) to improve readability. Image smoothing: Simulating higher-resolution color transitions in gradient areas of photos (such as the sky) to reduce graininess.

[0077] In some embodiments, the second branch performs fast processing: the second branch uses a lightweight network (such as ShuffleNetV2) to extract pixel-level features (such as text outlines, image block regions) and outputs low-dimensional feature maps (such as original resolution features).

[0078] In some embodiments, the physical pixel capabilities of the laser printer (such as 300dpi rasterization) are directly matched to avoid overcomputation and ensure rapid output for low-demand tasks such as draft mode and black and white documents.

[0079] In some embodiments, the pre-identification layer has a fault-tolerant mechanism: if the pre-identification layer makes a mistake (such as misjudging a low-resolution image as requiring sub-pixel analysis), the first branch will automatically downgrade to pixel-level processing through a feature confidence threshold (such as Hessian matrix response value < 0.1) to avoid invalid calculations.

[0080] When the printer does not support subpixel-level parameters (such as insufficient laser pulse frequency adjustment range), it automatically converts the subpixel features extracted by the first branch into software simulation effects (such as bilinear interpolation sharpening) to ensure that the task can be completed.

[0081] In some embodiments, the pre-identification layer incorporates multimodal feature fusion: in addition to the visual features of the image, the pre-identification layer can fuse file metadata (such as high-priority tags in PDFs) and user historical preferences (such as mandatory high-definition processing of contract documents) to comprehensively determine whether sub-pixel analysis is required.

[0082] For documents with mixed content (such as PDFs containing tables and photos), weights are assigned by region (e.g., 70% weight for the table region + 30% weight for the photo region) to comprehensively determine whether to enable the sub-pixel branch. The sub-pixel features extracted by the first branch can be correlated with printer hardware capabilities (e.g., upper limit of laser dot density, fixing temperature range) to generate achievable super-resolution parameters. For example, if the printer supports a maximum of 1200 dpi, but the sub-pixel features require 1600 dpi, the feature map is compressed proportionally to an equivalent accuracy of 1200 dpi. For high-contrast areas (e.g., text-to-background contrast > 80%), the intensity of sub-pixel analysis is reduced to avoid overexposure due to excessive enhancement.

[0083] In some embodiments, the second branch integrates pixel-level feature templates of common document types (such as A4 contract templates and ID card scan templates), and accelerates the direct output of features through template matching, thereby reducing network inference time.

[0084] In some embodiments, the pre-recognition layer of the AI ​​model identifies the color richness and / or contour complexity of the image; when at least one of the color richness and the contour complexity meets a preset condition, it is determined that the sub-pixel level feature recognition needs to be performed.

[0085] In some embodiments, the pre-recognition layer of the AI ​​model analyzes the imaging features of the target file, extracts two core indicators—color richness and contour complexity—and determines whether to initiate sub-pixel level feature recognition based on preset conditions. The specific process is as follows:

[0086] Color richness refers to the diversity of color variations in an image, quantified by the entropy value of a color histogram. The calculation formula is: ;in, Let be the proportion of the i-th color in the image. This represents the total number of color categories (e.g., N=2563 in RGB space).

[0087] In some embodiments, . This is the color threshold, and its value can be, for example, 0.4, 0.5, or 0.6, or a value between 0 and 1. Sub-pixel analysis is required to determine high color richness.

[0088] In some embodiments, high color richness files (such as photos, posters) typically require multi-color toner collaborative control (such as CMYK four-color overlay) of laser printers, and sub-pixel level feature recognition can optimize toner distribution in color transition areas.

[0089] In some embodiments, contour complexity detection may include: contour complexity refers to the degree of tortuosity and density of edges in an image, measured by the standard deviation of edge gradient magnitudes. The calculation steps include:

[0090] Edge detection is performed on the image to obtain the edge intensity map G(x,y); the standard deviation of the global gradient magnitude is calculated: ;in The average gradient magnitude. This represents the total number of edge pixels. Preset conditions: If the gradient threshold is greater than or equal to the gradient threshold, the contour determination complexity is high, requiring sub-pixel analysis.

[0091] In some embodiments, documents with high contour complexity (such as dense curves in engineering drawings or handwritten signatures) require high-precision beam positioning of the laser printer (such as ±0.01mm scan offset control), and sub-pixel-level feature recognition can optimize edge sharpening effects.

[0092] In some embodiments, if the document type is a contract / invoice (identified via metadata or OCR), subpixel analysis is forced to ensure the clarity of small text (such as monetary figures).

[0093] In summary, by accurately distinguishing the types of printed materials, appropriate printing parameters are used. For example, for photos / posters: high color richness triggers subpixel analysis, optimizes the micro-mixing effect of CMYK toner, and reduces color banding.

[0094] For example, regarding engineering drawings or handwritten signatures: high profile complexity (e.g., ≥15) Trigger subpixel analysis to enhance the edge sharpness of fine lines at the 0.1mm level and avoid the staircase effect of laser printers.

[0095] In some embodiments, to reduce unnecessary computational load, plain text documents (such as the body of a PDF contract): with low color richness and low outline complexity, directly enter the second branch, skipping subpixel calculations, thus improving processing speed.

[0096] In some embodiments, detail enhancement instructions output by the subpixel branch (such as an edge sharpening weight of 80%) are directly correlated with the printer's variable aperture adjustment (adjusting the laser beam focusing range) and toner ejection precision (controlling micron-level ink droplet size). The rapid processing results of the pixel branch (such as low-resolution rasterized data) are matched with the printer's draft mode to achieve high-speed printing and low ink volume.

[0097] In some embodiments, the threshold color threshold and / or gradient threshold can be dynamically adjusted based on the printer hardware status. For example, if the remaining toner cartridge level is <20%, the color threshold can be increased (e.g., from 6.0 to 7.0) to reduce high-ink-consumption subpixel analysis tasks.

[0098] If a high-temperature environment is detected (the temperature of the fixing unit is greater than the temperature threshold, for example, the temperature threshold can be 180℃ or 200℃), the outline complexity threshold is reduced (e.g., the gradient threshold is reduced from 15 to 12) to prioritize text clarity.

[0099] In some embodiments, the judgment results of preset conditions are overridden by combining file metadata (such as medical image tags) and user preferences (such as high precision priority). For example, even if the color threshold and gradient threshold are not exceeded, if the file type is a medical CT image, subpixel analysis is forced to ensure accurate restoration of bone / blood vessel edges.

[0100] In some embodiments, the first branch extracts sub-pixel level features of the image, including:

[0101] The first branch extracts sub-pixel level color values ​​from the image based on an interpolation algorithm, and uses a gradient algorithm to determine sub-pixel level color gradient values ​​based on the color values.

[0102] The first branch extracts pixel-level contours in the image based on an edge detection algorithm and extracts sub-pixel-level contour features based on a sub-pixel interpolation algorithm.

[0103] In some embodiments, subpixel-level color values ​​are extracted using interpolation algorithms (such as bilinear interpolation) to recover subtle color transitions lost due to pixel discretization in imaging (such as the gradation of the sky and skin tone transitions in a photograph). By combining gradient algorithms to calculate color gradient values, color edges (such as the boundary between text and background) are precisely located, enabling laser printers to achieve subpixel-level color compensation when CMYK toner is mixed, reducing color banding and graininess. After extracting pixel-level contours using edge detection algorithms (such as Canny), the edge positions are refined using subpixel interpolation algorithms (such as refining a 1-pixel wide line to 0.5 pixels), and subpixel-level contour features (such as edge curvature and direction) are extracted. This technology allows laser printers to enhance edge sharpness through high-precision laser positioning when printing small text (such as 6pt Song typeface), avoiding stair-step effects and improving text readability.

[0104] In some embodiments, the first branch extracts sub-pixel level features of the image, including:

[0105] The first branch extracts sub-pixel level features at different scales of the imaging based on multiple convolutional kernels of different sizes;

[0106] The first branch focuses on the subpixel-level features based on an attention mechanism to obtain subpixel-level color features and subpixel-level contour features.

[0107] In some embodiments, multi-scale sub-pixel features in the image are extracted in parallel using convolutional kernels of different sizes (e.g., 3×3, 5×5, 7×7), which can simultaneously capture local details (e.g., the tips of text strokes) and global structure (e.g., large areas of color in an image). This multi-scale feature fusion effectively solves the problem of detail loss or background interference at a single scale, and is particularly suitable for complex documents (e.g., mixed documents containing tables and photos). In some embodiments, attention mechanisms (e.g., SE modules or CBAM) are introduced to focus on key sub-pixel regions in the image (e.g., monetary figures in a contract, dimensions in engineering drawings) and suppress irrelevant background (e.g., paper texture, stains). This technology enables laser printers to prioritize high-priority areas when resources are limited, improving printing efficiency while ensuring the clarity of key content and reducing misidentification or blurring caused by background interference.

[0108] In summary, with interpolation and edge detection as its core, it focuses on the physical accuracy of color transition and edge sharpening, and is suitable for scenarios with high requirements for color and line accuracy (such as photo printing and engineering drawings).

[0109] With multi-scale convolution and attention as its core, it focuses on adaptive feature extraction in complex scenarios and is suitable for mixed content documents or scenarios that require priority processing of key information (such as contracts and invoices).

[0110] Both technologies have overcome the physical resolution limitations of laser printer hardware through algorithm optimization, achieving a synergistic upgrade of hardware capabilities and software algorithms, and significantly improving print quality and efficiency.

[0111] In some embodiments, the second branch extracts pixel-level features of the image, including:

[0112] The second branch extracts pixel-level features of the image based on convolutional layers;

[0113] The second branch extracts contour and color features from the pixel-level features based on the pooling layer.

[0114] This embodiment extracts pixel-level features of the image through a lightweight network structure of convolutional layers and pooling layers. Combined with the hardware characteristics of the laser printer, it achieves efficient and stable generation of basic printing parameters.

[0115] Local perception in convolutional layers: Pixel neighborhood features (such as text edges and color block distribution) are extracted by sliding 3×3 or 5×5 convolutional kernels. The computational complexity is much lower than that of interpolation / gradient algorithms for sub-pixel branches. Millisecond-level feature extraction can be achieved in low-resolution documents (such as 300dpi scans) or plain text files (such as PDF contracts), significantly improving task processing speed.

[0116] Global compression of pooling layers: Max pooling or average pooling is used to compress the feature map size (e.g., 2×2 pooling reduces the feature map size by 50%), preserving key contours (e.g., text skeletons) and color clustering information (e.g., background color blocks), reducing the computational load of subsequent parameter decisions.

[0117] The extracted pixel-level features are directly mapped to the physical pixel control capabilities of the laser printer (such as 300dpi rasterization imaging). The generated basic printing parameters (such as DPI and color mode) are fully compatible with the printer hardware, avoiding the waste of resources caused by high-precision calculations in subpixel branches, and ensuring the stability and speed of low-demand tasks such as draft mode and fast printing.

[0118] For low-complexity documents (such as solid color labels and simple charts), the pooling layer can effectively filter out noise interference (such as paper texture), output concise feature maps, reduce invalid toner ejection in laser printers, reduce energy consumption, and extend drum life.

[0119] A dynamic convolution kernel mechanism is introduced into the convolutional layers, adjusting the kernel size in real time based on the local complexity of the image (such as edge density and texture variations). For example, a 3×3 convolution kernel is used for text regions (high edge density) to accurately capture stroke edges; a 5×5 convolution kernel is used for background regions (low texture) to expand the receptive field and merge redundant information. This reduces redundant computation (such as over-analysis of simple backgrounds), improves feature extraction efficiency, and ensures the preservation of details in complex content, making it suitable for documents with mixed content (such as PDFs containing tables and text). Multi-scale pooling fusion: balancing global and local features, for example, combining max pooling (preserving salient features) and average pooling (smoothing noise) to design multi-scale pooling layers. For example, the first pooling layer uses max pooling to extract the text skeleton; the second pooling layer uses average pooling to compress background color block information. In this way, the generated feature map contains both salient contours and global color distribution, allowing the printer to maintain text clarity (such as key contract clauses) and background consistency (such as the background color of a flyer) even in draft mode. The pooled feature map is binarized (e.g., contour features are converted into a 0 / 1 matrix) to directly generate the control instruction set for the laser printer. For example, 1 corresponds to a high ink volume area (e.g., text strokes); 0 corresponds to a low ink volume area (e.g., background). This skips floating-point calculations in traditional parameter decision-making and accelerates print job startup through hardware-friendly binary instructions, making it particularly suitable for embedded laser printers (e.g., portable label printers).

[0120] In some embodiments, determining printing parameters for printing the target file based on at least two of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file includes: obtaining a preference matrix corresponding to the preference information, a state matrix corresponding to the printing parameters currently supported by the laser printer, and a feature matrix corresponding to the first feature; and determining the printing parameters based on the mapping relationship associated with the preference matrix, the state matrix, and the feature matrix.

[0121] In some embodiments, by using matrix modeling and dynamic mapping, user preferences, hardware status, and file characteristics are decoupled into quantifiable and computable parameter factors. This not only solves the problem of cumbersome manual adjustment of traditional printer parameters and waste of hardware resources, but also provides a standardized solution for complex scenarios such as mixed content printing and cross-device collaboration, which has significant engineering application value.

[0122] In some embodiments, the method further includes: when the amount of change of at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target document satisfies a first condition, adaptively implementing cross-page printing parameter switching during page turning of different pages of the target document; and / or, when the amount of change of at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target document satisfies a second condition, implementing cross-region printing parameter switching in different regions of a page of the target document.

[0123] This embodiment monitors changes in preference information, printer status, and document characteristics, and sets dynamic thresholds (first condition / second condition) to achieve cross-page parameter switching (e.g., different parameters for the cover and inner pages) or cross-region parameter switching (e.g., different parameters for text and image areas on the same page) during the printing process of the target document. The specific rules are as follows:

[0124] (1) Switching printing parameters across pages (meeting the first condition): Changes in preference information, user modification of printing settings (such as switching from draft mode to high quality mode); or, changes in printer status, consumables remaining amount is lower than the remaining threshold, or fixing temperature is abnormal; or, changes in document characteristics, or page type switching (such as jumping from a plain text page to a mixed text and graphics page), or sudden changes in color complexity (such as jumping from a grayscale page to a full-color page).

[0125] The threshold for the first condition is: the change in a single factor is greater than or equal to the preset benchmark value (e.g., the change in user preference weight is greater than or equal to 30%, and the change in consumable balance is greater than or equal to 20%); or the change in a combination of multiple factors reaches the joint threshold (e.g., the change in user preference is 10% and the printer temperature increases by 15%).

[0126] During page transitions (e.g., from page 1 to page 2), new parameters are preloaded through the printer cache to achieve seamless switching. For example, page 1 is the main body of the contract (plain text, user-preferred draft mode), printed at 600dpi+ in black and white; page 2 is the contract cover (including a color image of the company logo), and upon detecting a sudden change in color complexity (change ΔH_c=4.5>threshold 4.0), it automatically switches to 1200dpi+ CMYK four-color printing.

[0127] In some embodiments, cross-regional printing parameter switching is performed when the second condition is met.

[0128] Local feature changes in the document: color gradients in different areas within the same page (e.g., gradient difference in text areas < threshold 10, gradient difference in image areas > threshold 15), and contour density (e.g., edge density in table areas < threshold 8, edge density in photo areas > threshold 20).

[0129] The threshold for the second condition is: the change in regional features exceeds the local benchmark value (e.g., the color difference at the boundary between text and image is ≥2.0 or the gradient difference is ≥12); or the difference between global features and regional features is significant.

[0130] In some embodiments, parameters such as laser power and toner ejection volume are dynamically adjusted as the printhead scans different areas. For example, in a contract page, the text area (small Song typeface) uses high contrast parameters (laser power 50mW, resolution 600dpi); the image area (company logo) detects a sudden change in color gradient (Δσ_G=22>threshold 15) and automatically switches to high precision mode (laser power 60mW, resolution 1200dpi).

[0131] In another embodiment, a method for adaptively adjusting laser printer parameters includes: obtaining a user preference matrix. The feature matrix corresponding to the first feature is identified through an AI model. Real-time monitoring of the laser printer's status, for example, real-time monitoring of the laser printer's hardware status matrix. . It can be used to represent laser power deviation, photosensitive drum charge attenuation rate, fixing temperature fluctuation rate, scanning position error, etc. It is fused based on dynamic weighting coefficients α, β, γ. , , Generate optimized printing parameters.

[0132] In some embodiments, Table 3 includes a matrix characterizing the current state of the laser printer.

[0133]

[0134] Table 3

[0135] in, Calculated dynamically from the attenuation sensitivity model. The value of i can be 1, 2, 3, or 4. For example, (e.g., when the photosensitive drum wears down, the attenuation sensitivity is increased). (Weights). For example, fusion based on dynamic weight coefficients α, β, γ. , , The optimized printing parameters can be generated using the following functional relationship: .in, These are the optimized printing parameters. : Hardware state matrix (real-time sensor data). α, β, γ: Dynamic weight coefficients, which can be dynamically adjusted through reinforcement learning, or they can be static parameters. : Parameter optimization function, for example, This can represent a neural network mapping layer. In some embodiments, the dynamic weight coefficients are updated through a reinforcement learning model, and the reward function is: ,in, For print quality, For printing energy consumption, This refers to the status changes of the laser printer.

[0136] Parameter adjustments between pages will be performed when any of the following conditions are met:

[0137] Content type switching: The content characteristics of two consecutive pages change significantly (e.g., page 1 is plain text, page 2 is an RGB image).

[0138] User preference changes: Users can adjust their preference settings in real time during the printing process via the APP (such as switching from standard mode to high quality mode).

[0139] Hardware status warning: The sensor detects a sudden change in the status of a critical component (such as a sudden increase in the temperature of the photosensitive drum by 5°C or a fluctuation in laser power exceeding ±5%).

[0140] Substrate type change: The paper thickness / type detector detects a change in the media properties of the next page (e.g., switching from 80g plain paper to 200g coated paper).

[0141] Preload analysis: When printing page N, pre-parse the PCL / PS data stream of page N+1 and identify content characteristics (text / image / mixed type) through an AI model.

[0142] Parameter switching strategy: If the difference in characteristics between adjacent pages exceeds a threshold (e.g., SSIM (structural similarity) < 0.6), a print parameter reset is triggered. Specifically, new print parameters are loaded during the page gap. For user preference changes: The current transmission queue is immediately interrupted, and a parameter update command is inserted. Hardware coordination ensures: Laser power calibration and fixing temperature adjustment are performed during the page gap to avoid switching parameters within the print line.

[0143] Switching delay control: Cross-page parameter switching must be completed within the page turning interval (e.g., <50ms) to avoid affecting throughput.

[0144] Within a single page, the following conditions must be met simultaneously: significant differences in regional characteristics: for example, text areas are adjacent to image areas (such as a title + photo combination), and the resolution difference is greater than a specified value, such as 200 dpi.

[0145] There is a coexistence of gradient regions and RGB image regions (requiring different halftone algorithms).

[0146] Limited local scanning capabilities of laser printers: The scanning motor cannot maintain high-precision positioning when moving at high speeds (the local scanning speed needs to be reduced).

[0147] Region segmentation and parameter mapping: An AI model is used to divide a printed page into several regions (such as a title area, image area, and background area), and feature labels are generated for each region.

[0148] Dynamic laser control: Based on the feature label, the printing parameters are adaptively switched by region during the scanning process.

[0149] Adaptive halftone switching algorithm:

[0150] Text area: Use raster binarization to preserve sharp edges;

[0151] Gradient region: Switch to error diffusion algorithm to prevent color banding.

[0152] Dynamic scanning speed reduction: Automatically reduces the scanning speed in handwritten text areas with high curvature. Formula:

[0153] ;in, This is the adjusted scan rate. The basic scan rate for printing the current page. The curvature of the current text stroke. The maximum curvature of the stroke.

[0154] In some embodiments, hardware provides real-time feedback: an integrated photoelectric sensor detects the actual imaging effect at the end of the scan line. If ghosting / blurring is detected at the boundary of the region, compensation is immediately triggered: the number of scan repetitions at the boundary is increased, and the laser power at the boundary is finely adjusted.

[0155] In some embodiments, when the parameter differences between adjacent areas within the same page are too large (e.g., laser power difference > 30%), a transition buffer (2-3 scan lines) is inserted, and linear interpolation is used to smooth the transition parameters.

[0156] like Figure 3 As shown, this disclosure provides an adaptive adjustment device for printing parameters of a laser printer, comprising:

[0157] The first determining module 3110 is used to determine the preference information of the user printing the target file; the preference information is used to determine the printing parameters of the first scale at least.

[0158] The recognition module 3120 is used to recognize a first feature of the target file using an artificial intelligence (AI) model; the first feature includes pixel-level features and / or sub-pixel-level features; the first feature is used to determine printing parameters at a second scale; the second scale is smaller than the first scale;

[0159] The second determining module 3130 is used to determine the printing parameters for printing the target file based on at least two of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file.

[0160] In summary, the laser printer printing parameter adaptive adjustment device provided in this embodiment can realize the laser printer printing parameter adaptive adjustment method provided in any of the foregoing embodiments.

[0161] Combination Figure 4 As shown, this application embodiment provides an electronic device including a processor 10 and a memory 11. Optionally, the device may further include a communication interface 12 and a bus 9. The processor 10, communication interface 12, and memory 11 can communicate with each other via the bus 9. The communication interface 12 can be used for information transmission. The processor 10 can call logical instructions in the memory 11 to execute the adaptive adjustment method for printing parameters of the laser printer described in the above embodiment. This electronic device can be a controller for a laser printer.

[0162] Furthermore, the logical instructions in the aforementioned memory 11 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0163] The memory 11, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this application. The processor 10 executes functional applications and data processing by running the program instructions / modules stored in the memory 11, that is, it implements the adaptive adjustment method of printing parameters of the laser printer in the above embodiments.

[0164] The memory 11 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 11 may include high-speed random access memory and may also include non-volatile memory.

[0165] This electronic device can be used as an edge controller, central controller, or edge control terminal, etc.

[0166] This application provides a computer program product, which includes a computer program stored on a storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the above-described adaptive adjustment method for printing parameters of a laser printer.

[0167] The aforementioned computer-readable storage medium may be a transient computer-readable storage medium or a non-transitory computer-readable storage medium.

[0168] The technical solutions of this application embodiment can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes one or more instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in this application embodiment. The aforementioned storage medium can be a non-transitory storage medium, including various media capable of storing program code such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks, or it can be a transient storage medium.

[0169] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0170] The embodiments or examples disclosed in this application are not exhaustive, but merely illustrative of some embodiments or examples, and are not intended to limit the scope of protection of this disclosure. Unless contradictory, each step in a particular embodiment or example can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment or example can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment or example can be arbitrarily interchanged. Furthermore, optional methods or examples in a particular embodiment or example can be arbitrarily combined; moreover, embodiments or examples can be arbitrarily combined. For example, some or all steps of different embodiments or examples can be arbitrarily combined, and a particular embodiment or example can be arbitrarily combined with optional methods or examples of other embodiments or examples.

[0171] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0172] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0173] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0174] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0175] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for adaptive adjustment of printing parameters of a laser printer, characterized in that, The method includes: Determine the printing user's preference information for the target file; the preference information is used to determine at least the printing parameters for the first scale; The target file is identified using an artificial intelligence (AI) model; the first feature includes pixel-level features and / or sub-pixel-level features; the first feature is used to determine printing parameters at a second scale; the second scale is smaller than the first scale; the pixel-level features include: edges, textures, and / or text stroke widths in the image; the sub-pixel-level features include: halftone dot density distribution; the printing parameters at the first scale include at least one of the following: page settings; color type settings; the printing parameters at the second scale include at least one of the following: laser beam wavelength, laser pulse repetition frequency, laser scanning speed, laser dot density, laser dot power, and number of printing laser groups; Based on the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file, the printing parameters for printing the target file are determined, including: obtaining the preference matrix corresponding to the preference information, the state matrix corresponding to the printing parameters currently supported by the laser printer, and the feature matrix corresponding to the first feature; and determining the printing parameters based on the mapping relationship related to the preference matrix, the state matrix, and the feature matrix, including: fusing dynamic weight coefficients α, β, and γ. , , The optimized printing parameters are generated; The preference matrix; The feature matrix corresponding to the first feature; The hardware state matrix in the state matrix is ​​related to at least one of the following: laser power deviation, photosensitive drum charge attenuation rate, fixing temperature fluctuation rate, and scanning position error of the laser printer. During the printing of the target document, when the change in at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target document satisfies the first condition, the printing parameter switching across pages is adaptively implemented during page turning of different pages of the target document. During the printing of a target document, when the change in at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target document satisfies a second condition, cross-regional printing parameter switching is implemented in different areas of a page of the target document.

2. The method according to claim 1, characterized in that, The step of determining the printing parameters for printing the target file based on the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file further includes at least one of the following: The AI ​​model dynamically optimizes printing parameters, including, when a table area is identified, activating the vector printing mode of laser printing, and / or, when a handwritten signature is identified, switching the resolution mode and magnifying the local area of ​​the handwritten signature. The AI ​​model dynamically adjusts the laser power based on the printing load, including reducing the laser power during single-page printing to extend the drum life. And / or, activate the residual heat cache of the fuser component during batch printing.

3. The method according to claim 1 or 2, characterized in that, The process of using an artificial intelligence (AI) model to identify the pixel-level features of the target file includes: The AI ​​model's pre-recognition layer identifies the image of the target file and determines whether the image needs to undergo sub-pixel level feature recognition. When the imaging requires sub-pixel level feature recognition, the imaging is input into the first branch of the AI ​​model; The first branch extracts sub-pixel level features of the image; When the imaging does not require subpixel-level feature recognition, the imaging is input into the second branch of the AI ​​model; The second branch extracts pixel-level features of the image.

4. The method according to claim 3, characterized in that, The pre-recognition layer of the AI ​​model identifies the image of the target file and determines whether the image needs to undergo sub-pixel level feature recognition, including: The pre-recognition layer of the AI ​​model identifies the color richness and / or contour complexity of the image; When at least one of the color richness and the contour complexity meets a preset condition, it is determined that the sub-pixel level feature recognition needs to be performed.

5. The method according to claim 3, characterized in that, The first branch extracts sub-pixel level features of the image, including: The first branch extracts sub-pixel level color values ​​from the image based on an interpolation algorithm, and uses a gradient algorithm to determine sub-pixel level color gradient values ​​based on the color values. The first branch extracts pixel-level contours in the image based on an edge detection algorithm and extracts sub-pixel-level contour features based on a sub-pixel interpolation algorithm.

6. The method according to claim 3, characterized in that, The first branch extracts sub-pixel level features of the image, including: The first branch extracts sub-pixel level features at different scales of the imaging based on multiple convolutional kernels of different sizes; The first branch focuses on the subpixel-level features based on an attention mechanism to obtain subpixel-level color features and subpixel-level contour features.

7. The method according to claim 3, characterized in that, The second branch extracts pixel-level features of the image, including: The second branch extracts pixel-level features of the image based on convolutional layers; The second branch extracts contour and color features from the pixel-level features based on the pooling layer.

8. A laser printer's adaptive adjustment device for printing parameters, characterized in that, include: The first determining module is used to determine the printing preference information of the target file user; The preference information is used at least to determine the printing parameters for the first scale; The recognition module is used to recognize a first feature of the target file using an artificial intelligence (AI) model; the first feature includes pixel-level features and / or sub-pixel-level features; the first feature is used to determine printing parameters at least at the second scale; The second scale is smaller than the first scale; The pixel-level features include: edges, textures, and / or text stroke widths in the image; the sub-pixel-level features include: halftone dot density distribution; the printing parameters at the first scale include at least one of the following: page settings; color type settings; the printing parameters at the second scale include at least one of the following: laser beam wavelength, laser pulse repetition frequency, laser scanning speed, laser dot density, laser dot power, and number of printing laser groups; The second determining module is used to determine printing parameters for printing the target file based on the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file. This includes: obtaining a preference matrix corresponding to the preference information, a state matrix corresponding to the printing parameters currently supported by the laser printer, and a feature matrix corresponding to the first feature; and determining the printing parameters based on the mapping relationship related to the preference matrix, state matrix, and feature matrix, including: fusing dynamic weight coefficients α, β, and γ. , , The optimized printing parameters are generated; The preference matrix; The feature matrix corresponding to the first feature; The hardware state matrix in the state matrix is ​​related to at least one of the following: laser power deviation, photosensitive drum charge attenuation rate, fixing temperature fluctuation rate, and scanning position error of the laser printer. The second determining module is further configured to, during the printing of the target file, adaptively switch printing parameters across pages during page turning in the target file when the change in at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file satisfies a first condition; and during the printing of the target file, when the change in at least one of the preference information, the printing parameters currently supported by the laser printer, and the first feature of the target file satisfies a second condition, switch printing parameters across regions in different areas of a page of the target file.

Images