Image adaptive optimization processing method and system for laser printing output

Abstract

The application relates to the technical field of image data processing, in particular to an image adaptive optimization processing method and system for laser printing output, which comprises the following steps: performing multi-scale feature analysis and fusion on an input original scanning image to obtain a fused feature distribution mapping matrix; performing adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain an enhanced contrast image; performing edge feature extraction on the enhanced contrast image by using an improved multi-direction edge detection algorithm to obtain an edge-enhanced edge feature image; performing adaptive local texture analysis and enhancement on the edge feature image to obtain a texture-enhanced image; and performing printing adaptability optimization based on the texture-enhanced image to obtain a final optimized output image. According to the technical scheme, the image quality printed and output by a laser printing device is significantly improved from multiple image processing dimensions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image data processing, and particularly relates to an image adaptive optimization processing method and system for laser printing output. BACKGROUND

[0002] With the rapid development of digital image processing and printing technology, image optimization as a key technology to improve visual quality and printing effect plays an increasingly important role in document processing, commercial printing and other fields. Especially in the laser printing scene, due to the inherent characteristics of the printing equipment and the requirement of black and white binary output, how to maintain image details while achieving high-quality printing output has become a technical problem to be solved in this field.

[0003] In the prior art, image optimization methods based on color space conversion are commonly used. For example, a Chinese invention patent with application number CN202411930713.8 discloses a multimedia image optimization method, which realizes brightness enhancement and contrast adjustment through HSI (Hue, Saturation, Intensity, i.e. hue, saturation, brightness / intensity) color space conversion, and combines edge enhancement technology to improve image quality. However, this kind of method mainly aims at display optimization of color images, and has obvious limitations in the laser printing scene: first, although color space conversion can improve the overall visual effect of the image, it fails to fully consider the output characteristics of the printing equipment, resulting in distortion of the optimized image when printing; second, traditional edge enhancement methods often use fixed enhancement parameters, which are difficult to adapt to the local feature differences of different image contents, and are prone to cause over-enhancement or insufficient enhancement problems; third, these methods lack targeted optimization strategies when dealing with image texture details, and it is difficult to ensure good texture performance in black and white binary output.

[0004] In addition, when performing image binarization processing, the prior art usually adopts global threshold or simple local threshold method, which is difficult to balance the overall performance and local details of the image, and is prone to cause quality problems such as detail loss and edge blur in printing output. At the same time, due to the lack of adaptive optimization mechanism for printing resolution, the existing method is difficult to ensure stable output effect on different printing equipment.

[0005] Therefore, how to develop an image processing method that can comprehensively consider image enhancement, texture optimization and printing adaptability to meet the special needs of laser printing scene and achieve high-quality black and white printing output has important practical significance and application value. SUMMARY

[0006] In order to solve the technical problem that the actually printed image quality is poor in the existing image optimization scheme for laser printing output, the purpose of the present application is to provide an image adaptive optimization processing method and system for laser printing output, and the technical scheme is as follows:

[0007] The present application provides an image adaptive optimization processing method for laser printing output, which comprises:

[0008] Performing multi-scale feature analysis and fusion on the input original scan image to obtain a fused feature distribution mapping matrix;

[0009] Performing adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain an enhanced contrast image;

[0010] Performing edge feature extraction on the enhanced contrast image using an improved multi-directional edge detection algorithm to obtain an edge-enhanced edge feature image; the improved multi-directional edge detection algorithm performs omnidirectional edge feature detection through a directional gradient operator group in a radial-tangential coordinate system, and introduces an adaptive weight function based on local structure similarity to suppress noise and maintain edge features;

[0011] Performing adaptive local texture analysis and enhancement on the edge feature image to obtain a texture-enhanced image;

[0012] Performing printing adaptability optimization based on the texture-enhanced image to obtain a final optimized output image.

[0013] Further, the multi-scale feature analysis and fusion on the input original scan image to obtain a fused feature distribution mapping matrix comprises:

[0014] Determining a multi-scale window sequence corresponding to a target pixel coordinate in the original scan image; the multi-scale window sequence comprises a plurality of target scale windows;

[0015] Determining a first mapping value of the gray scale statistical feature distribution mapping matrix of the target scale window at the target pixel coordinate;

[0016] Determining a weight coefficient of the target scale window and a local correlation coefficient at the target pixel coordinate;

[0017] Using the first mapping value, the local correlation coefficient and the weight coefficient to determine the fused feature distribution mapping matrix corresponding to the multi-scale window sequence.

[0018] Further, the first mapping value of the gray scale statistical feature distribution mapping matrix of the target scale window at the target pixel coordinate comprises:

[0019] determine an offset gray value of the target pixel coordinate in the original scan image based on a horizontal offset and a vertical offset in the multi-scale window sequence;

[0020] determine a first mapping value of a gray value of the target pixel coordinate in the target scale window based on the offset gray value and a gray variance in the target scale window;

[0021] obtain the gray statistical feature distribution mapping matrix of the target scale window based on the first mapping value of the gray value of each pixel coordinate.

[0022] Further, the adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain the enhanced contrast image, comprising:

[0023] determine a contrast adjustment factor and a local threshold parameter of the target pixel coordinate in the original scan image;

[0024] obtain an adaptive contrast mapping function based on the contrast adjustment factor, the local threshold parameter, and a second mapping value of the fused feature distribution mapping matrix at the target pixel coordinate;

[0025] obtain the enhanced contrast image by performing a nonlinear gray mapping based on a third mapping value of the adaptive contrast mapping function at the target pixel coordinate and a local brightness compensation parameter term of the target pixel coordinate.

[0026] Further, the determination of the contrast adjustment factor and the local threshold parameter of the target pixel coordinate in the original scan image comprises:

[0027] determine a mean value and a standard deviation of the fused feature distribution mapping matrix in a neighborhood window where the target pixel coordinate is located;

[0028] determine a horizontal offset and a vertical offset of the fused feature distribution mapping matrix in the neighborhood window where the target pixel coordinate is located;

[0029] obtain a target offset coordinate of the target pixel coordinate based on the horizontal offset and the vertical offset, and determine a fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate;

[0030] determine the contrast adjustment factor of the target pixel coordinate in the original scan image based on the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate, the mean value, and the standard deviation;

[0031] determine the local threshold parameter of the target pixel coordinate in the original scan image based on the mean value and a preset local contrast gain factor.

[0032] Further, the improved multi-directional edge detection algorithm is used to extract edge features from the enhanced contrast image to obtain an edge-enhanced edge feature image, including:

[0033] Based on the constructed multi-directional gradient operator group, a response value of a multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image is determined;

[0034] The response values at each direction angle are fused to obtain a fifth mapping value of the fused gradient amplitude map at the target pixel coordinate;

[0035] Based on the fifth mapping value of the fused gradient amplitude map at each pixel coordinate, a morphological enhancement operation is performed to obtain the edge-enhanced edge feature image.

[0036] Further, the response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image is determined based on the constructed multi-directional gradient operator group, including:

[0037] Based on the radial coordinate and the tangential coordinate of the operator in the multi-directional gradient operator group, a shifted pixel value of the shifted pixel coordinate of the target pixel coordinate in the enhanced contrast image combined with the radial coordinate and the tangential coordinate of the operator is determined;

[0038] The response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image is determined using the shifted pixel value and the adaptive weight function based on the local structure similarity of the shifted pixel coordinate.

[0039] Further, the adaptive local texture analysis and enhancement of the edge feature image is performed to obtain a texture-enhanced image, including:

[0040] Based on the constructed adaptive analysis window, a local texture feature value at the target pixel coordinate of the edge feature image is determined;

[0041] The texture-enhanced image is obtained by performing texture enhancement on the target pixel coordinate using the local texture feature value.

[0042] Further, the printing adaptability optimization based on the texture-enhanced image is performed to obtain a final optimized output image, including:

[0043] The corrected pixel value and the corrected image of the target pixel coordinate in the texture-enhanced image are obtained by performing gray scale correction on the texture-enhanced image using the correction factor corresponding to the printing resolution of the target printing device;

[0044] Based on the horizontal offset and the vertical offset in the constructed adaptive binarization window, an offset pixel value at the offset pixel coordinate of the target pixel coordinate in the corrected image is obtained based on the horizontal offset and the vertical offset;

[0045] An adaptive threshold at a target pixel coordinate in the corrected image is determined using an offset pixel value in the corrected image, and a final optimized output image is obtained based on the corrected pixel value and the adaptive threshold.

[0046] The application also provides an image adaptive optimization processing system for laser printing output, which is used to implement the image adaptive optimization processing method for laser printing output as any one of the above; the system comprises:

[0047] A feature fusion module is configured to perform multi-scale feature analysis and fusion on the input original scan image to obtain a fused feature distribution mapping matrix.

[0048] A contrast enhancement module is configured to perform adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain an enhanced contrast image.

[0049] An edge enhancement module is configured to perform edge feature extraction on the enhanced contrast image using an improved multi-directional edge detection algorithm to obtain an edge-enhanced edge feature image; the improved multi-directional edge detection algorithm performs omnidirectional edge feature detection through a directional gradient operator group in a radial-tangential coordinate system and introduces an adaptive weight function based on local structure similarity to suppress noise and maintain edge features.

[0050] A texture enhancement module is configured to perform adaptive local texture analysis and enhancement on the edge feature image to obtain a texture-enhanced image.

[0051] A printing optimization module is configured to perform printing adaptability optimization based on the texture-enhanced image to obtain a final optimized output image.

[0052] The application has the following beneficial effects:

[0053] (1) The improved multi-directional edge detection algorithm is used to perform edge feature extraction in the application, a multi-directional gradient operator group with radial and tangential characteristics is constructed, and omnidirectional edge structure information in an image can be captured.

[0054] (2) The application realizes intelligent enhancement of image texture through adaptive local texture analysis and enhancement processing. By constructing an adaptive analysis window and introducing a Gaussian weight function, the analysis range can be flexibly adjusted according to the local features of the image, ensuring the accuracy of texture feature extraction. In the texture enhancement process, an adaptive calculation method based on local statistical features is used, so that the enhancement effect can better adapt to the changes in image content. By adaptively fusing edge features and texture features, the texture details of the image are enhanced, and the edge definition is maintained, effectively avoiding the edge blurring problem caused by traditional texture enhancement methods.

[0055] (3) The application designs a complete print adaptability optimization scheme for the special needs of the laser printing scene. By introducing a gray scale correction mechanism, adaptive adjustment can be made according to the output characteristics of different printing devices, ensuring that the image can obtain the best output effect under different printing resolutions. An adaptive local threshold algorithm is used in the binarization processing link, which considers the gray scale distribution characteristics of the local area, realizing more accurate black and white conversion. This optimization strategy not only compensates for the inherent characteristic deviation of the printing device, but also enhances the level of image expression, so that the final printed output not only maintains the original detail features, but also ensures high definition and level.

[0056] The technical scheme of the application comprehensively and significantly improves the image quality printed by the laser printing device from the above multiple aspects. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0058] Figure 1 A step flow chart of an image adaptive optimization processing method for laser printing output provided by an embodiment of the application;

[0059] Figure 2 A detailed flow chart of step S1 in an image adaptive optimization processing method for laser printing output provided by an embodiment of the application;

[0060] Figure 3 A detailed flow chart of step S12 in an image adaptive optimization processing method for laser printing output provided by an embodiment of the application;

[0061] Figure 4A refinement flowchart of step S2 in a laser printing output-oriented image adaptive optimization processing method provided by an embodiment of the present application;

[0062] Figure 5 A refinement flowchart of step S21 in a laser printing output-oriented image adaptive optimization processing method provided by an embodiment of the present application;

[0063] Figure 6 A refinement flowchart of step S3 in a laser printing output-oriented image adaptive optimization processing method provided by an embodiment of the present application;

[0064] Figure 7 A refinement flowchart of step S4 in a laser printing output-oriented image adaptive optimization processing method provided by an embodiment of the present application;

[0065] Figure 8 A refinement flowchart of step S5 in a laser printing output-oriented image adaptive optimization processing method provided by an embodiment of the present application;

[0066] Figure 9 A laser printing output-oriented image adaptive optimization processing method-related image processing enhancement progressive result schematic diagram of an embodiment of the present application: (a) input example image, (b) edge enhancement image, (c) texture enhancement image, (d) output example image;

[0067] Figure 10 A hardware running environment structure schematic diagram of a laser printing output-oriented image adaptive optimization processing device-related embodiment of the present application;

[0068] Figure 11 A framework structure schematic diagram of a laser printing output-oriented image adaptive optimization processing system-related embodiment of the present application. DETAILED DESCRIPTION

[0069] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined purposes, the following describes a laser printing output-oriented image adaptive optimization processing method according to the present application, its specific implementation, structure, features and effects in detail in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0071] The application provides a specific scheme of an image adaptive optimization processing method for laser printing output.

[0072] Embodiment one:

[0073] For the image adaptive optimization processing method for laser printing output, refer to Figure 1 , which shows a step flowchart of the image adaptive optimization processing method for laser printing output provided by an embodiment of the application.

[0074] The image adaptive optimization processing method for laser printing output comprises:

[0075] Step S1: performing multi-scale feature analysis and fusion on the input original scanning image to obtain a fused feature distribution mapping matrix;

[0076] Specifically, refer to Figure 2 , the step S1 comprises:

[0077] Step S11: determining a multi-scale window sequence corresponding to a target pixel coordinate in the original scanning image; the multi-scale window sequence comprises a plurality of target scale windows;

[0078] Step S12: determining a first mapping value of a gray scale statistical feature distribution mapping matrix of the target scale window at the target pixel coordinate;

[0079] More specifically, refer to Figure 3 , the step S12 comprises:

[0080] Step S121: determining an offset gray scale value of the target pixel coordinate under horizontal and vertical offsets in the original scanning image based on a horizontal offset and a vertical offset in the multi-scale window sequence;

[0081] Step S122: determining the first mapping value of the gray scale of the target pixel coordinate in the gray scale statistical feature distribution mapping matrix of the target scale window based on the offset gray scale value and a gray scale variance in the target scale window;

[0082] Step S123: obtaining the gray scale statistical feature distribution mapping matrix of the target scale window based on the first mapping value of the gray scale of each pixel coordinate.

[0083] Step S13: determining a weight coefficient of the target scale window and a local correlation coefficient at the target pixel coordinate;

[0084] Step S14: determining the fused feature distribution mapping matrix corresponding to the multi-scale window sequence based on the first mapping value, the local correlation coefficient and the weight coefficient.

[0085] In this embodiment, a multi-scale analysis window sequence is first constructed and grayscale statistics are performed, specifically as follows:

[0086] Obtain the original scanned image P, such as Figure 9 As shown in (a), and in its pixel coordinates (m, n) (where n represents the target pixel coordinates), Figure 9 An analysis window of size (2k+1)×(2k+1) is constructed at (not shown in the figure), where k is the window scale parameter, k∈[1,3], to obtain the multi-scale window sequence W={W1,W2,W3}. In this embodiment, the window sizes corresponding to W1, W2 and W3 are 3×3, 5×5 and 7×7, respectively.

[0087] Calculate the gray-level statistical features within each target scale window (referring to any scale window) k to obtain its feature distribution mapping matrix (gray-level statistical feature distribution mapping matrix) H, specifically:

[0088]

[0089] Among them, H k (m,n) is the first mapping value of the feature distribution mapping matrix at any k-th scale at pixel coordinates (m,n) (representing the target pixel coordinates, referring to any pixel coordinate); i and j are the horizontal and vertical offsets within the multi-scale window sequence, respectively;

[0090] P(m+i,n+j) is the gray value (offset gray value) of the original scanned image at pixel coordinates (m+i,n+j); V k (m,n) represents the gray-level variance within the k-th scale window; α is the target variance threshold, which is 100 in this embodiment; e is the natural constant; Hk(m,n) based on the coordinates of each pixel constitutes the gray-level statistical feature distribution mapping matrix H.

[0091] The specific steps for constructing a multi-scale feature fusion mapping are as follows:

[0092]

[0093] Where Q(m,n) is the second mapping value of the fused feature distribution mapping matrix Q at pixel coordinates (m,n); β k Let be the weight coefficient for the k-th scale, representing the contribution of different scales, satisfying... In this embodiment, β1, β2, and β3 are 0.5, 0.3, and 0.2, respectively; γ k (m,n) represents the local correlation coefficient at pixel coordinates (m,n) at the k-th scale, specifically:

[0094]

[0095] where max(P(m, n)) and min(P(m, n)) are the maximum and minimum intensity values of the original scanned image within the scale window at pixel coordinate (m, n).

[0096] The above implementation process can effectively capture the local gray scale statistical features of the image at different scales by constructing a sequence of multi-scale analysis windows to analyze and fuse the features of the original scanned image. By calculating the mean and variance of the gray scale of the local region at each scale and combining the exponential weighted adaptive mapping mechanism, the feature extraction process has good response capability to both local details and overall structure of the image. At the same time, the local correlation coefficient is introduced to adaptively fuse the features at different scales, which not only maintains the structural integrity at a larger scale, but also retains the detail information at a smaller scale. The multi-scale feature fusion strategy used in this step can effectively balance the local detail preservation and global feature expression of the image, providing a reliable feature basis for subsequent image enhancement processing.

[0097] Step S2, performing adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain an enhanced contrast image;

[0098] Specifically, referring to Figure 4 , the step S2 comprises:

[0099] Step S21, determining a contrast adjustment factor and a local threshold parameter at the target pixel coordinate in the original scanned image;

[0100] More specifically, referring to Figure 5 , the step S21 comprises:

[0101] Step S211, determining the mean and standard deviation of the fused feature distribution mapping matrix within the neighborhood window where the target pixel coordinate is located;

[0102] Step S212, determining the horizontal and vertical offsets of the fused feature distribution mapping matrix within the neighborhood window where the target pixel coordinate is located;

[0103] Step S213, obtaining the target offset coordinate of the target pixel coordinate using the horizontal and vertical offsets, and determining the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate;

[0104] Step S214, determining the contrast adjustment factor at the target pixel coordinate in the original scanned image using the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate, the mean and the standard deviation;

[0105] Step S215, determining the local threshold parameter at the target pixel coordinate in the original scan image by using the mean value and the preset local contrast gain factor.

[0106] Step S22, obtaining the adaptive contrast mapping function by using the contrast adjustment factor, the local threshold parameter and the second mapping value of the fused feature distribution mapping matrix at the target pixel coordinate;

[0107] Step S23, performing the nonlinear gray mapping to obtain the enhanced contrast image by using the third mapping value of the adaptive contrast mapping function at the target pixel coordinate and the local brightness compensation parameter term at the target pixel coordinate.

[0108] In the embodiment, the adaptive contrast mapping function needs to be constructed, specifically:

[0109]

[0110] wherein, is the third mapping value of the adaptive contrast mapping function at the pixel coordinate (m, n); δ(m, n) is the contrast adjustment factor at the pixel coordinate (m, n); ρ(m, n) is the local threshold parameter at the pixel coordinate (m, n);

[0111] The adaptive parameters of the local region are calculated, specifically:

[0112]

[0113] ;

[0114] wherein, and are the horizontal and vertical offset amounts in the 5x5 neighborhood window of the target pixel coordinate (m, n) respectively; is the fourth mapping value of the fused feature distribution mapping matrix Q at the target offset pixel coordinate ; and are the mean value and the standard deviation of the fused feature distribution mapping matrix Q in the 5x5 neighborhood window at the pixel coordinate (m, n) respectively; ε is a positive number for preventing division by zero, and ε = 10 -6 ; τ1 and τ2 are weight coefficients, and in the embodiment, τ1 and τ2 are 0.8 and 0.4 respectively; Φ(m, n) is the local contrast gain factor, specifically:

[0115]

[0116] Wherein, φ1, φ2 and φ3 are adjustment parameters, φ1, φ2 and φ3 in the embodiment are respectively 1.1, 0.5 and 0.2; Ψ(m, n) is a local gradient intensity, which in the embodiment is specifically:

[0117]

[0118] Wherein, Q(m+1, n), Q(m+1, n), Q(m, n+1) and Q(m, n-1) are respectively mapping values of the fused feature distribution mapping matrix at pixel coordinates

[0119] (m+1, n), (m-1, n), (m, n+1) and (m, n-1) of the enhanced contrast image;

[0120] Then, a non-linear gray scale mapping is performed to obtain an enhanced contrast image , which is specifically:

[0121]

[0122] Wherein, is a pixel value of the enhanced contrast image at pixel coordinate (m, n) (further obtaining the overall enhanced contrast image); P(m, n) is a pixel value of the original scanning image P at pixel coordinate (m, n); ω(m, n) is a local brightness compensation parameter term (or a function term) at pixel coordinate (m, n), which is specifically:

[0123]

[0124] Wherein, is a mean value of a 5x5 neighborhood window of the original scanning image P at pixel coordinate (m, n); ζ1 and ζ2 are weight coefficients, ζ1 and ζ2 in the embodiment are respectively 0.6 and 0.4; is a weight value of the Gaussian weight matrix at offset , which is specifically:

[0125]

[0126] The above implementation process realizes dynamic contrast enhancement of the image by constructing an adaptive contrast mapping function, and the advantage is that the enhancement degree can be adaptively adjusted according to the local features of the image. By introducing a non-linear mapping mechanism, the image dark area detail performance can be effectively improved, and over-enhancement and saturation of the bright area can be avoided. In the parameter calculation process, the local region statistical features and gradient information are used to adaptively determine the enhancement parameters, which not only ensures the local adaptability of the enhancement effect, but also maintains the visual coordination of the image as a whole.

[0127] Step S3, edge feature extraction is performed on the enhanced contrast image using an improved multi-directional edge detection algorithm to obtain an edge-enhanced edge feature image;

[0128] The improved multi-directional edge detection algorithm performs omnidirectional edge feature detection through a set of directional gradient operators in a radial-tangential coordinate system, and introduces an adaptive weight function based on local structural similarity to suppress noise and maintain edge features.

[0129] Specifically, referring to Figure 6 , the step S3 comprises:

[0130] Step S31, based on the constructed multi-directional gradient operator set, the response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image is determined.

[0131] More specifically, the step S31 comprises:

[0132] Based on the radial and tangential coordinates of the operators in the multi-directional gradient operator set, the shifted pixel value of the shifted pixel coordinate of the target pixel coordinate in the enhanced contrast image combined with the radial and tangential coordinates of the operators is determined.

[0133] The shifted pixel value and the adaptive weight function based on local structural similarity of the shifted pixel coordinate are used to determine the response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image.

[0134] Step S32, the response values under each direction angle are fused to obtain the fifth mapping value of the fused gradient amplitude map at the target pixel coordinate.

[0135] Step S33, based on the fifth mapping value of the fused gradient amplitude map at each pixel coordinate, a morphological enhancement operation is performed to obtain an edge-enhanced edge feature image.

[0136] In this embodiment, a multi-directional gradient operator set needs to be constructed, specifically:

[0137]

[0138] wherein, is the value of the gradient operator with direction angle θ at coordinate , θ ∈ [0°, 360°), and the sampling interval is 45°; and are the radial and tangential coordinates of the operator respectively; η1 and η2 are the scale parameters of the radial and tangential directions respectively, which are 1.5 and 0.8 respectively in this embodiment; the multi-directional gradient response map is calculated, specifically:

[0139]

[0140] wherein, is the response value of the gradient response map with a direction angle of θ at the pixel coordinate (m, n) in the enhanced contrast image; is the enhanced contrast image shifts the pixel coordinate (m, n) by a value of to obtain a shifted pixel coordinate (m', n'); is an adaptive weight function at the pixel coordinate (m', n'), specifically,

[0141]

[0142] wherein, is a local structure similarity at the pixel coordinate (m', n'), in the embodiment, specifically,

[0143]

[0144] wherein, is the enhanced contrast image at the pixel coordinate (m, n);

[0145] Then, the multi-direction gradient responses are fused and the principal direction features are extracted, specifically,

[0146]

[0147]

[0148] wherein, is the fused gradient amplitude map at the pixel coordinate (m, n); is the corresponding principal direction angle map at the pixel coordinate (m, n);

[0149] Finally, a morphological enhancement operation is performed, specifically,

[0150]

[0151] wherein, is the target pixel coordinate for edge enhancement, that is, the result after edge enhancement is performed on each pixel coordinate, and thus the enhanced edge feature image , for example,​​​​Figure 9 (b) the value (1 or 0) at the pixel coordinate ; is a morphological dilation operation; is a morphological erosion operation.

[0152] The above implementation process uses an improved multi-directional edge detection algorithm to extract edge features, which has the advantage of being able to capture edge structural information in the image in all directions. By constructing a multi-directional gradient operator group with radial and tangential characteristics, this step can effectively detect edge features in different directions, avoiding the limitations of the directionality of traditional edge detection algorithms. In the gradient response calculation process, the adaptive weight function introduced can dynamically adjust the detection sensitivity according to the local image structure characteristics, which not only improves the accuracy of edge detection, but also enhances the algorithm's resistance to noise.

[0153] Step S4, performing adaptive local texture analysis and enhancement on the edge feature image to obtain a texture-enhanced image;

[0154] Specifically, referring to Figure 7 , the step S4 comprises:

[0155] Step S41, determining a local texture feature value at a target pixel coordinate of the edge feature image based on the constructed adaptive analysis window;

[0156] Step S42, performing texture enhancement on the target pixel coordinate using the local texture feature value to obtain a texture-enhanced image.

[0157] In this embodiment, an adaptive analysis window needs to be constructed, specifically:

[0158]

[0159] wherein, is a local analysis result at the pixel coordinate of the edge-enhanced edge feature image; is a local analysis window radius, which is 3 in this embodiment; is a value of the edge-enhanced edge feature image at the pixel coordinate ; and are horizontal and vertical offsets within the local analysis window, respectively; is a Gaussian weight function, which in this embodiment is specifically:

[0160]

[0161] Further, the local texture feature is calculated, specifically:

[0162]

[0163] in, pixel coordinates in the edge feature image Local texture feature values ​​at the location; for In pixel coordinates The mean value within the central local analysis window;

[0164] Then, texture enhancement is performed, specifically:

[0165]

[0166] in, Enhance images with texture ,like Figure 9 As shown in (c), at pixel coordinates The pixel value at that location (1 or 0); The texture enhancement factor is 0.6 in this embodiment.

[0167] The above implementation process effectively enhances the texture detail of images through adaptive local texture analysis and enhancement. By constructing an adaptive analysis window and introducing a Gaussian weighting function, this step can flexibly adjust the analysis range based on local image features, ensuring the accuracy of texture feature extraction while avoiding abrupt changes between local regions. During texture feature calculation, an adaptive calculation method based on local statistical features is employed, enabling texture feature extraction to better adapt to changes in image content and achieving reasonable enhancement effects for different types of texture regions.

[0168] Step S5: Perform printability optimization based on the texture enhancement image to obtain the final optimized output image.

[0169] Specifically, please refer to Figure 8 Step S5 includes:

[0170] Step S51: Using the correction factor corresponding to the printing resolution of the target printing device, perform grayscale correction on the texture enhancement image to obtain the corrected pixel value of the target pixel coordinates and the corrected image in the texture enhancement image.

[0171] Step S52: Based on the horizontal and vertical offsets within the constructed adaptive binarization window, determine the offset pixel value at the offset pixel coordinates of the target pixel in the corrected image at the horizontal and vertical offsets.

[0172] Step S53: Determine the adaptive threshold at the target pixel coordinates in the corrected image using the offset pixel value in the corrected image, and obtain the final optimized output image based on the corrected pixel value and the adaptive threshold.

[0173] Based on the above embodiments, in this embodiment, grayscale correction is required to adapt to the printer's output characteristics, specifically as follows:

[0174]

[0175] in, For the corrected image In pixel coordinates The value at that location (corrected pixel value); The printer calibration factor is 1.2 in this embodiment; For the target printing device's print resolution The corresponding correction factor is 1.4 in this embodiment;

[0176] Adaptive binarization is performed to obtain the final optimized output image, specifically:

[0177] Calculate local thresholds based on adaptive binarization windows:

[0178]

[0179] in, pixel coordinates in the corrected image Adaptive threshold at the location; To achieve an adaptive binarization window radius, it is set to 7 in this embodiment; For the corrected image Offset pixel coordinates The offset pixel value at that location; The offset is fixed; in this embodiment, it is 5. and These are the horizontal and vertical offsets within the adaptive binarization window, respectively;

[0180] Binarization processing is performed, specifically as follows:

[0181]

[0182] Among them, F For the final optimized output image F, such as Figure 9 As shown in (d), its pixel coordinates The value at the specified location (1 or 0) is used to obtain the final optimized output image F.

[0183] Embodiment two:

[0184] The embodiment of the present application also provides an image adaptive optimization processing device for laser printing output. The device can be a laser printer, a computer, a server or a combination of multiple data analysis and calculation devices.

[0185] As shown in Figure 10 , the hardware environment structure of the image adaptive optimization processing device for laser printing output is shown in Figure 10 .

[0186] As shown in Figure 10 , the image adaptive optimization processing device for laser printing output can include a processor 1001 such as a CPU, a network interface 1004, a user interface 1003, a memory 1005 and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display, an input unit such as a control panel, and the optional user interface 1003 can also include a standard wired interface and a wireless interface. The network interface 1004 can optionally include a standard wired interface and a wireless interface (such as a WIFI interface). The memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory) such as a disk memory. The memory 1005 can also be an independent storage device from the aforementioned processor 1001. The memory 1005 as a computer storage medium can include an image adaptive optimization processing program.

[0187] Those skilled in the art can understand that Figure 10 the hardware structure shown in does not constitute a limitation on the device, and can include more or fewer components than shown, or combine certain components, or different component arrangements.

[0188] Figure 10 Continuing to refer to Figure 10 , the memory 1005 as a computer readable storage medium can include an operating system, a user interface module, a network communication module and an image adaptive optimization processing program.

[0189] In Figure 10 , the network communication module is mainly used to connect the server and can communicate data with the server; and the processor 1001 can call the image adaptive optimization processing program stored in the memory 1005 and execute the steps in the above various embodiments.

[0190] Based on the hardware structure of the image adaptive optimization processing device for laser printing output, various embodiments of the image adaptive optimization processing method for laser printing output of the present application are implemented.

[0191] In addition, the present application also provides an image adaptive optimization processing system for laser printing output (referred to as "image adaptive optimization processing system" for short), please refer to Figure 11 , the image adaptive optimization processing system for laser printing output comprises:

[0192] The feature fusion module A10 is used for multi-scale feature analysis and fusion of the input original scan image, and a fused feature distribution mapping matrix is obtained.

[0193] The contrast enhancement module A20 is used for adaptive contrast enhancement based on the fused feature distribution mapping matrix, and an enhanced contrast image is obtained.

[0194] The edge enhancement module A30 is used for edge feature extraction of the enhanced contrast image using an improved multi-directional edge detection algorithm, and an edge-enhanced edge feature image is obtained. The improved multi-directional edge detection algorithm performs omnidirectional edge feature detection through a directional gradient operator group in a radial-tangential coordinate system, and introduces an adaptive weight function based on local structure similarity to suppress noise and maintain edge features.

[0195] The texture enhancement module A40 is used for adaptive local texture analysis and enhancement of the edge feature image, and a texture-enhanced image is obtained.

[0196] The print optimization module A50 is used for print adaptability optimization based on the texture-enhanced image, and a final optimized output image is obtained.

[0197] Further, the feature fusion module A10 is also used for:

[0198] Determining a multi-scale window sequence corresponding to a target pixel coordinate in the original scan image; the multi-scale window sequence comprises a plurality of target scale windows.

[0199] Determining a first mapping value of the gray scale statistical feature distribution mapping matrix of the target scale window at the target pixel coordinate;

[0200] Determining a weight coefficient of the target scale window and a local correlation coefficient at the target pixel coordinate;

[0201] Using the first mapping value, the local correlation coefficient and the weight coefficient, a fused feature distribution mapping matrix corresponding to the multi-scale window sequence is determined.

[0202] Further, the feature fusion module A10 is also used for:

[0203] determine, based on the horizontal offset and the vertical offset in the sequence of multi-scale windows, an offset gray value of the target pixel coordinate in the original scan image under the horizontal and vertical offset;

[0204] determine, based on the offset gray value and a gray variance in the target scale window, a first mapping value of a gray value of the gray statistical feature distribution mapping matrix of the target scale window at the target pixel coordinate;

[0205] obtain the gray statistical feature distribution mapping matrix of the target scale window based on the first mapping value of the gray value at each pixel coordinate.

[0206] Further, the contrast enhancement module A20 is further configured to:

[0207] determine a contrast adjustment factor and a local threshold parameter of the target pixel coordinate in the original scan image;

[0208] obtain an adaptive contrast mapping function using the contrast adjustment factor, the local threshold parameter, and a second mapping value of the fused feature distribution mapping matrix at the target pixel coordinate;

[0209] perform a nonlinear gray mapping to obtain an enhanced contrast image using a third mapping value of the adaptive contrast mapping function at the target pixel coordinate and a local brightness compensation parameter term of the target pixel coordinate.

[0210] Further, the contrast enhancement module A20 is further configured to:

[0211] determine a mean value and a standard deviation of the fused feature distribution mapping matrix within a neighborhood window in which the target pixel coordinate is located;

[0212] determine a horizontal offset and a vertical offset of the fused feature distribution mapping matrix within the neighborhood window in which the target pixel coordinate is located;

[0213] obtain a target offset coordinate of the target pixel coordinate using the horizontal offset and the vertical offset, and determine a fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate;

[0214] determine a contrast adjustment factor of the target pixel coordinate in the original scan image using the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinate, the mean value, and the standard deviation;

[0215] determine a local threshold parameter of the target pixel coordinate in the original scan image using the mean value and a preset local contrast gain factor.

[0216] Further, the edge enhancement module A30 is further configured to:

[0217] Determine a response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image based on the constructed multi-directional gradient operator set;

[0218] Fuse the response values at the various direction angles to obtain a fifth mapping value of the fused gradient magnitude map at the target pixel coordinate;

[0219] Perform a morphological enhancement operation based on the fifth mapping values of the fused gradient magnitude map at the various pixel coordinates to obtain an edge-enhanced edge feature image.

[0220] Further, the edge enhancement module A30 is further configured to:

[0221] Determine a shifted pixel value of a shifted pixel coordinate of the target pixel coordinate in the enhanced contrast image in combination with the radial coordinate and the tangential coordinate of the operator based on the radial coordinate and the tangential coordinate of the operator in the multi-directional gradient operator set;

[0222] Determine the response value of the multi-directional gradient response map at the target pixel coordinate in the enhanced contrast image using the shifted pixel value and an adaptive weight function based on the local structural similarity of the shifted pixel coordinate.

[0223] Further, the texture enhancement module A40 is further configured to:

[0224] Determine a local texture feature value at the target pixel coordinate of the edge feature image based on the constructed adaptive analysis window;

[0225] Perform texture enhancement on the target pixel coordinate using the local texture feature value to obtain a texture-enhanced image.

[0226] Further, the print optimization module A50 is further configured to:

[0227] Perform gray scale correction on the texture-enhanced image using a correction factor corresponding to a print resolution of the target print device to obtain a corrected pixel value and a corrected image of the target pixel coordinate in the texture-enhanced image;

[0228] Determine an offset pixel value at an offset pixel coordinate of the target pixel coordinate in the corrected image based on a horizontal offset and a vertical offset within the constructed adaptive binarization window;

[0229] Determine an adaptive threshold value at the target pixel coordinate in the corrected image using the offset pixel value in the corrected image, and obtain a final optimized output image based on the corrected pixel value and the adaptive threshold value.

[0230] The embodiment of the present application facing the image adaptive optimization processing system of laser printing output is basically the same as the above-mentioned embodiment of the image adaptive optimization processing method facing the image adaptive optimization processing system of laser printing output, and will not be repeated here.

[0231] In addition, the present application also provides a computer readable storage medium. The computer readable storage medium of the present application stores an image adaptive optimization processing program, wherein the image adaptive optimization processing program is executed by a processor to realize the steps of the image adaptive optimization processing method facing the image adaptive optimization processing system of laser printing output.

[0232] The method realized by the image adaptive optimization processing program can refer to the embodiments of the image adaptive optimization processing method facing the image adaptive optimization processing system of laser printing output, and will not be repeated here.

[0233] It should be noted that the above-mentioned embodiment of the present application is only for description, and does not represent the advantages and disadvantages of the embodiment. The processes described in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0234] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments.

[0235] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0236] The above-mentioned is only the preferred embodiment of the present application, and does not limit the protection scope of the present application. Any equivalent structure / method transformation made according to the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the protection scope of the present application.

Claims

1. An adaptive optimization processing method for images output by laser printing, characterized in that, The method includes: Multi-scale feature analysis and fusion are performed on the input raw scan image to obtain the fused feature distribution mapping matrix; Adaptive contrast enhancement is performed based on the fused feature distribution mapping matrix to obtain an enhanced contrast image. An improved multi-directional edge detection algorithm is used to extract edge features from the enhanced contrast image to obtain an edge feature image with enhanced edges. The improved multi-directional edge detection algorithm performs omnidirectional edge feature detection through a group of directional gradient operators in the radial-tangential coordinate system, and introduces an adaptive weight function based on local structural similarity to suppress noise and preserve edge features. Adaptive local texture analysis and enhancement are performed on the edge feature image to obtain a texture-enhanced image; Printability optimization is performed based on texture-enhanced images to obtain the final optimized output image; The improved multi-directional edge detection algorithm is used to extract edge features from the enhanced contrast image to obtain an edge feature image with enhanced edges, including: Based on the constructed multi-directional gradient operator group, the response value of the multi-directional gradient response map at the target pixel coordinates in the enhanced contrast image is determined; By fusing the response values ​​from various directional angles, the fifth mapping value of the fused gradient magnitude map at the target pixel coordinates is obtained; Based on the fifth mapping value of the fused gradient magnitude map at each pixel coordinate, morphological enhancement operation is performed to obtain an edge feature image with edge enhancement. The method for determining the response value of the multi-directional gradient response map at the target pixel coordinates in the enhanced contrast image based on the constructed multi-directional gradient operator set includes: Based on the radial and tangential coordinates of the operators in the multi-directional gradient operator group, the shifted pixel value of the target pixel coordinates in the enhanced contrast image is determined by combining the radial and tangential coordinates of the operators. Using the shifted pixel value and the shifted pixel coordinates, an adaptive weighting function based on local structural similarity is used to determine the response value of the multi-directional gradient response map at the target pixel coordinates in the enhanced contrast image.

2. The image adaptive optimization processing method for laser printing output according to claim 1, characterized in that, The step of performing multi-scale feature analysis and fusion on the input original scanned image to obtain the fused feature distribution mapping matrix includes: Determine a multi-scale window sequence corresponding to the target pixel coordinates in the original scanned image; the multi-scale window sequence includes multiple target scale windows. Determine the first mapping value of the gray-scale statistical feature distribution mapping matrix of the target scale window at the target pixel coordinates; Determine the weighting coefficients of the target scale window and the local correlation coefficients at the target pixel coordinates; Using the first mapping value, the local correlation coefficient, and the weight coefficient, the feature distribution mapping matrix corresponding to the fused multi-scale window sequence is determined.

3. The image adaptive optimization processing method for laser printing output according to claim 2, characterized in that, The first mapping value of the gray-level statistical feature distribution mapping matrix of the target scale window at the target pixel coordinates includes: Based on the horizontal and vertical offsets in the multi-scale window sequence, the offset gray values ​​of the target pixel coordinates in the original scanned image under the horizontal and vertical offsets are determined. Using the offset gray value and the gray variance in the target scale window, the first mapping value of the gray statistical feature distribution mapping matrix of the target scale window at the target pixel coordinates is determined; Based on the first mapping value of gray level at each pixel coordinate, the gray level statistical feature distribution mapping matrix of the target scale window is obtained.

4. The image adaptive optimization processing method for laser printing output according to claim 1, characterized in that, The adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain the enhanced contrast image includes: Determine the contrast adjustment factor and local threshold parameter at the target pixel coordinates in the original scanned image; An adaptive contrast mapping function is obtained by using the contrast adjustment factor, the local threshold parameter, and the second mapping value of the fused feature distribution mapping matrix at the target pixel coordinates. The enhanced contrast image is obtained by performing nonlinear grayscale mapping using the third mapping value of the adaptive contrast mapping function at the target pixel coordinates and the local brightness compensation parameter at the target pixel coordinates.

5. The image adaptive optimization processing method for laser printing output according to claim 4, characterized in that, The process of determining the contrast adjustment factor and local threshold parameter at the target pixel coordinates in the original scanned image includes: Determine the mean and standard deviation of the fused feature distribution mapping matrix within the neighborhood window of the target pixel coordinates; Determine the horizontal and vertical offsets of the fused feature distribution mapping matrix within the neighborhood window of the target pixel coordinates; The target offset coordinates of the target pixel coordinates are obtained using the horizontal and vertical offsets, and the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinates is determined. The contrast adjustment factor at the target pixel coordinates in the original scanned image is determined by using the fourth mapping value of the fused feature distribution mapping matrix at the target offset coordinates, the mean, and the standard deviation. Using the mean and a preset local contrast gain factor, the local threshold parameter at the target pixel coordinates in the original scanned image is determined.

6. The image adaptive optimization processing method for laser printing output according to claim 1, characterized in that, The adaptive local texture analysis and enhancement of the edge feature image to obtain a texture-enhanced image includes: Based on the constructed adaptive analysis window, the local texture feature values ​​at the target pixel coordinates of the edge feature image are determined; The texture-enhanced image is obtained by performing texture enhancement at the target pixel coordinates using local texture feature values.

7. The image adaptive optimization processing method for laser printing output according to claim 1, characterized in that, The step of performing print adaptability optimization based on texture-enhanced images to obtain the final optimized output image includes: By using the correction factor corresponding to the printing resolution of the target printing device, grayscale correction is performed on the texture enhancement image to obtain the corrected pixel value and the corrected image of the target pixel coordinates in the texture enhancement image; Based on the horizontal and vertical offsets within the constructed adaptive binarization window, the offset pixel value is obtained by determining the offset pixel coordinates of the target pixel in the corrected image at the offset pixel coordinates of the horizontal and vertical offsets. An adaptive threshold is determined at the target pixel coordinates in the corrected image using the offset pixel value in the corrected image, and the final optimized output image is obtained based on the corrected pixel value and the adaptive threshold.

8. An image adaptive optimization processing system for laser printing output, characterized in that, The system is used to implement the image adaptive optimization processing method for laser printing output as described in any one of claims 1 to 7; the system includes: The feature fusion module is used to perform multi-scale feature analysis and fusion on the input raw scan image to obtain the fused feature distribution mapping matrix; The contrast enhancement module is used to perform adaptive contrast enhancement based on the fused feature distribution mapping matrix to obtain an enhanced contrast image. An edge enhancement module is used to extract edge features from an enhanced contrast image using an improved multi-directional edge detection algorithm to obtain an enhanced edge feature image. The improved multi-directional edge detection algorithm performs omnidirectional edge feature detection using a set of directional gradient operators in a radial-tangential coordinate system, and introduces an adaptive weight function based on local structural similarity to suppress noise and preserve edge features. It is also used to determine the response value of the multi-directional gradient response map at the target pixel coordinates in the enhanced contrast image based on the constructed multi-directional gradient operator set; fuse the response values ​​at various directional angles to obtain the fifth mapping value of the fused gradient magnitude map at the target pixel coordinates; perform morphological enhancement operations based on the fifth mapping value of the fused gradient magnitude map at each pixel coordinate to obtain the enhanced edge feature image; and further, it is used to determine the shifted pixel value of the target pixel coordinates in the enhanced contrast image by combining the radial and tangential coordinates of the operators in the multi-directional gradient operator set with the shifted pixel coordinates; and use the adaptive weight function based on local structural similarity of the shifted pixel value and the shifted pixel coordinates to determine the response value of the multi-directional gradient response map at the target pixel coordinates in the enhanced contrast image. The texture enhancement module is used to perform adaptive local texture analysis and enhancement on the edge feature image to obtain a texture-enhanced image; The print optimization module is used to perform print adaptability optimization based on texture enhancement images to obtain the final optimized output image.

Images