Digital processing system of natural wood grain texture images and colors based on machine vision

Abstract

The application provides a machine vision-based digital processing system for natural wood grain texture images and colors, comprising: an image acquisition unit for acquiring personification color information features; a color difference recognition and segmentation unit for recognizing natural defect types in a natural wood electronic original, including knots, dead knots, cracks, mineral lines and color change regions; the application effectively overcomes the influence of natural wood base color depth and surface roughness on inkjet color development by introducing an iteration formula for substrate reflectivity and roughness compensation coefficients, realizes accurate restoration on deep or uneven wood, and simulates the visual neglect characteristics of the human eye for texture regions through a weighted average color value algorithm, so that the automatic detection system can objectively and accurately evaluate the color difference of non-uniform surfaces, avoid misjudgment, and realize full-process automation from defect recognition, color calculation, iteration printing to quality detection, greatly improving the efficiency of wood color modification process implementation.

Description

Technical Field

[0001] This invention relates to the fields of wood processing and digital image processing technology, and in particular to a digital processing system for natural wood veneer texture images and colors based on machine vision. Background Technology

[0002] With the development of the home furnishing industry, consumers are increasingly demanding the natural aesthetic appeal of wood products. Natural wood veneer, due to its unique texture and color, is widely used in high-end furniture and interior decoration. However, natural wood inevitably has natural defects such as knots, cracks, and color differences. Traditional methods of processing its electronic originals mostly rely on manual repair, which is inefficient and lacks realism.

[0003] In recent years, digital inkjet printing technology has been introduced into the field of wood repair. Existing image processing systems are mostly based on paper printing models, assuming that the substrate is white and has a smooth surface. Natural wood veneer, as a non-white substrate, has varying shades of base color and natural roughness and pores on its surface. This results in a huge difference between the color rendering effect after inkjet printing and the theoretical value, and problems such as "dark color rendering", "color drift" or "ink bleeding" are very likely to occur.

[0004] Furthermore, when assessing color difference after repair, traditional color difference measuring instruments (such as colorimeters) typically perform point measurements. Even if multiple point measurements are taken and the average is calculated, the randomness of manual point selection and sampling means that for non-uniform surfaces like natural wood veneer with complex textures, simple averaging will include the brightness variations of the texture itself in the color difference calculation. This results in the measurement results failing to accurately reflect the color blending degree between the repaired area and the surrounding wood. Therefore, a digital processing system for natural wood veneer texture images and colors based on machine vision is proposed. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a digital processing system for natural wood veneer texture images and colors based on machine vision, in order to solve or alleviate the technical problems existing in the prior art, and at least provide a beneficial option.

[0006] The technical solution of this invention is implemented as follows: a digital processing system for natural wood veneer texture images and colors based on machine vision, comprising: An image acquisition unit is used to acquire anthropomorphic color information features; The color difference recognition and segmentation unit is used to identify the types of natural defects in the electronic original of natural wood veneer, including live knots, dead knots, cracks, mineral lines and color change areas, and to extract the boundary and texture features of the color difference areas; The color iteration correction unit is used to calculate the amount of color correction ink in the defect area for non-white substrates and non-smooth surfaces, and iteratively corrects the color through closed-loop control of printing detection to make the repair result close to the target texture. The weighted color difference measurement unit is used to evaluate the color difference of the repair results of non-uniform surfaces. It calculates the weighted average color value by constructing a weight matrix to avoid the influence of texture on color evaluation. The central control processor is used to coordinate the data flow and logical operations of the above-mentioned units.

[0007] In some embodiments, the color iteration correction unit performs the following steps: S1. Obtain the base color and establish a color overprinting model for a non-white substrate; S2. Calculate the initial printing ink volume based on the difference between the target color and the base color; S3. After performing the initial color correction, the color information of the printed result is acquired again through the image acquisition unit; S4. Calculate the deviation vector between the current result and the target color, calculate the corrected ink volume change based on the deviation vector, and proceed to the next printing iteration until the color difference value is less than the preset threshold and is jointly confirmed by humans, then terminate the iteration process.

[0008] In some embodiments, the calculation of establishing the color overprinting model for non-white substrates and correcting the ink volume increment uses the following formula: Let the target color vector be defined. Substrate background color vector , No. The actual color vector for this print is: ; Then the first Ink volume adjustment value required for the next iteration Calculated using the following formula: in, This is the ink volume conversion coefficient matrix. This is the substrate roughness compensation coefficient. The average reflectance of the substrate surface. The reflectance of a standard white substrate; No. The total ink volume for this print run is: .

[0009] In some embodiments, the weighted color difference measurement unit includes a weight matrix construction module, which assigns weight values ​​to pixels in the image based on the texture frequency features after image capture. Regions with high texture frequency or distinct edge features are assigned lower weight values; Pixels with low texture frequency or in flat areas are assigned higher weight values.

[0010] In some embodiments, the formula for calculating the weighted average color value is as follows: Let the set of pixels within the repair area be... pixel The chromaticity value is The weight coefficient of this pixel is ; The weighted average color value of the repaired area The calculation formula is: Among them, the weighting coefficient Determined by the following formula: In the formula, For pixels The gradient magnitude at a given point represents the degree of drastic change in texture. This is a texture-sensitive factor used to adjust the degree of influence of non-smooth surfaces on color difference measurements.

[0011] In some embodiments, the system further includes an ink droplet diffusion simulation module, used to simulate the physical diffusion behavior of ink on a non-smooth wood veneer surface before color iteration correction, generate a virtual print preview image, and assist in calculating the initial ink volume.

[0012] In some embodiments, the identification and automatic correction of natural defects in the electronic original of the natural wood veneer, including live knots, dead knots, cracks, mineral lines, and discoloration areas, do not alter the original characteristics.

[0013] In some embodiments, the color iteration correction unit works in conjunction with the weighted color difference measurement unit, when the weighted average color difference... If the color correction is below the set threshold, the system determines that the color correction is acceptable and stops iterative printing.

[0014] In some embodiments, an iterative approach is adopted to address the color rendering issue on non-white substrates. The system first identifies the wood veneer base color, introduces the substrate reflectivity parameter during printing, and uses a closed-loop feedback mechanism to re-acquire images after each print, calculate the deviation between the current color and the target color, and dynamically adjust the ink volume until the repair result visually and numerically approximates the target texture.

[0015] In some embodiments, for the problem of measuring non-uniform surfaces, the present invention proposes a weighted average color value calculation tool based on image capture. This tool assigns differentiated weights to image pixels: reducing the weights at texture edges and high-frequency regions, and increasing the weights in flat color regions, thereby filtering out the interference of texture structure on color evaluation and accurately calculating the true color difference perceived by the human eye.

[0016] The embodiments of the present invention have the following advantages due to the adoption of the above technical solutions: This invention effectively overcomes the influence of the dark base color and rough surface of natural wood veneer on inkjet color rendering by introducing an iterative formula for substrate reflectivity and roughness compensation coefficient. It achieves accurate reproduction on dark or uneven wood veneer and simulates the human eye's visual neglect of texture areas through a weighted average color value algorithm. This enables the automated detection system to objectively and accurately evaluate the color difference of non-uniform surfaces, avoiding misjudgments. Thus, it realizes full-process automation from defect identification, color calculation, iterative printing to quality inspection, and greatly improves the efficiency of wood veneer color change process.

[0017] The above overview is for illustrative purposes only and is not intended to be limiting in any way. Detailed Implementation

[0018] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention.

[0019] It is important to note that terms such as "first," "second," "symmetric," "array," "set in," and "set with" are used only to distinguish between descriptive and positional descriptions and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, features specified with terms such as "first" or "symmetric" may explicitly or implicitly include one or more of that feature; similarly, when the quantity of certain features is not limited by words such as "two" or "three," it should be noted that such features also explicitly or implicitly include one or more features.

[0020] In this invention, unless otherwise explicitly specified and limited, terms such as “installation,” “connection,” and “fixation” should be interpreted broadly; for example, they can be fixed connections, detachable connections, or integral moldings; they can be mechanical connections, direct connections, welding, or indirect connections through an intermediate medium; they can be internal connections between two components or the interaction between two components.

[0021] This invention provides a digital processing system for natural wood veneer texture images and colors based on machine vision, comprising: An image acquisition unit is used to acquire anthropomorphic color information features; The color difference recognition and segmentation unit is used to identify the types of natural defects in the electronic original of natural wood veneer, including live knots, dead knots, cracks, mineral lines and color change areas, and to extract the boundary and texture features of the color difference areas; The color iteration correction unit is used to calculate the amount of color correction ink in the defect area for non-white substrates and non-smooth surfaces, and iteratively corrects the color through closed-loop control of printing detection to make the repair result close to the target texture. The weighted color difference measurement unit is used to evaluate the color difference of the repair results of non-uniform surfaces. It calculates the weighted average color value by constructing a weight matrix to avoid the influence of texture on color evaluation. The central control processor is used to coordinate the data flow and logical operations of the above-mentioned units.

[0022] In this embodiment, the color iteration correction unit specifically performs the following steps: S1. Obtain the base color and establish a color overprinting model for a non-white substrate; S2. Calculate the initial printing ink volume based on the difference between the target color and the base color; S3. After performing the initial color correction, the color information of the printed result is acquired again through the image acquisition unit; S4. Calculate the deviation vector between the current result and the target color, calculate the corrected ink volume change based on the deviation vector, and proceed to the next printing iteration until the color difference value is less than the preset threshold and is jointly confirmed by humans, then terminate the iteration process.

[0023] In this embodiment, the following formula is used to establish the color overprinting model for non-white substrates and to calculate the ink volume increment: Let the target color vector be defined. Substrate background color vector , No. The actual color vector for this print is: ; Then the first Ink volume adjustment value required for the next iteration Calculated using the following formula: in, This is the ink volume conversion coefficient matrix. This is the substrate roughness compensation coefficient. The average reflectance of the substrate surface. The reflectance of a standard white substrate; No. The total ink volume for this print run is: .

[0024] In this embodiment, specifically, ΔInki+1 is the first The next iteration requires supplementing the ink volume vector (usually containing components of the four CMYK channels, and can be set to a maximum of 8 color vectors).

[0025] It is the target color vector, usually represented by the Lab color space, and is the average color of the normal wood surrounding the repair area.

[0026] : No. The actual color vector collected after the first printing.

[0027] This is an ink volume conversion coefficient matrix used to convert Lab color difference values ​​into ink droplet volumes or gray levels that the printer can recognize. This matrix needs to be obtained through printer color linearization calibration.

[0028] It is the substrate roughness compensation coefficient. Since the ink diffusion rate is different on non-smooth surfaces, this coefficient is used to correct the loss of color development efficiency caused by penetration.

[0029] It is the average reflectance of the base material color. This value is lower for dark wood veneers (such as walnut) and higher for light wood veneers (such as maple).

[0030] It is the reflectance of a standard white substrate (usually set as a reference value of 1.0 or 100%).

[0031] In this embodiment, specifically, the weighted color difference measurement unit includes a weight matrix construction module, which assigns weight values ​​to pixels in the image based on the texture frequency features after image capture. Regions with high texture frequency or distinct edge features are assigned lower weight values; Pixels with low texture frequency or in flat areas are assigned higher weight values.

[0032] In this embodiment, the specific formula for calculating the weighted average color value is as follows: Let the set of pixels within the repair area be... pixel The chromaticity value is The weight coefficient of this pixel is ; The weighted average color value of the repaired area The calculation formula is: Among them, the weighting coefficient Determined by the following formula: In the formula, For pixels The gradient magnitude at a given point represents the degree of drastic change in texture. This is a texture-sensitive factor used to adjust the degree of influence of non-smooth surfaces on color difference measurements.

[0033] In this embodiment, specifically, It is the final weighted average chromaticity value (luminance or chromaticity channel) of the repaired area.

[0034] Pixel The original chromaticity value at that location.

[0035] It is a pixel. The weighting coefficients.

[0036] It is a pixel. The image gradient magnitude at a given location is calculated as the absolute value of the grayscale difference between that pixel and its neighboring pixels, or as the result of the Sobel operator.

[0037] The greater the gradient, the more intense the texture (such as the edge of wood grain or the edge of a crack).

[0038] It is a texture sensitivity factor used to adjust the weight's sensitivity to texture.

[0039] In natural wood veneer, the wood grain lines (high gradient areas) themselves exhibit drastic color variations. Including them in the average value of color difference calculations would severely interfere with the judgment of the "tone" of the repair color.

[0040] By introducing gradient terms: At the texture lines Larger denominator → larger weight It gets smaller.

[0041] On a flat base color area Very small → weight Close to 1.

[0042] Calculated from this It can filter out high-frequency interference from the texture structure and truly reflect the "base color tone" of the wood, thereby achieving accurate color difference measurement of non-uniform surfaces.

[0043] In this embodiment, the system further includes an ink droplet diffusion simulation module, which is used to simulate the physical diffusion behavior of ink on a non-smooth wood veneer surface before color iteration correction, generate a virtual print preview image, and assist in calculating the initial ink volume.

[0044] In this embodiment, specifically, the identification and automatic correction of natural defects in the electronic original of natural wood veneer, including live knots, dead knots, cracks, mineral lines, and discoloration areas, do not change the original characteristics.

[0045] In this embodiment, specifically, the color iteration correction unit and the weighted color difference measurement unit work together, when the weighted average color difference... If the color correction is below the set threshold, the system determines that the color correction is acceptable and stops iterative printing.

[0046] In this embodiment, specifically, the weighted average color difference is calculated. This is not a simple two-point color difference, but a comprehensive color difference calculation that combines the "weighted average color value" in claim 5.

[0047] The color difference calculation formula (CIE Lab color difference formula) is shown below: in: It's a difference in brightness. It is the color difference between the red and green axes. It's a color difference between the yellow and blue axes.

[0048] In the formula , and It is not the original pixel difference, but the difference between the weighted average color value calculated based on claim 5 and the target color in claim 3.

[0049] The specific substitution relationships are as follows: Brightness difference Calculation: in, The weighted average brightness value calculated by the formula in claim 5 (i.e., the brightness after considering non-uniform texture suppression). The target brightness is as defined in claim 3.

[0050] Color difference and Calculation: Similarly, using the weight matrix in claim 5 Calculate the weighted average chromaticity value of the repaired area respectively. and \bar{b}_{measure}bˉmeasure: Then calculate the color difference: The mathematical expression for determining whether a system's decision is qualified is: in To set a threshold (e.g.) or This value can be set by software based on human visual tolerance or industry standards (such as CIELAB standards).

[0051] In this embodiment, specifically, in this system, the color iteration correction unit (claim 3) and the weighted color difference measurement unit (claim 5) constitute a closed-loop feedback loop. Due to the non-uniformity of the natural wood veneer surface, a single pixel comparison cannot accurately reflect the overall repair effect. Therefore, a color difference judgment algorithm based on weighted averaging is adopted.

[0052] The specific implementation steps are as follows: Data Acquisition: When the first After the color correction is completed in the second print, the image acquisition unit re-acquires the image of the repaired area.

[0053] Weight preprocessing: The system calls the weight matrix construction module to calculate the gradient weight of each pixel within the repair region. At this point, high-frequency information such as the edges of wood grain texture is given a low weight (e.g., 0.1), while repaired and smoothed areas are given a high weight (e.g., 0.9).

[0054] Weighted color value synthesis: The system uses the formula The "visual perception dominant color" of the repaired area is calculated. This calculation process actually performs weighted smoothing on the local color noise caused by uneven printing ink droplets or wood grain shadows, simulating the visual characteristics of the human eye that "ignore details and perceive the whole".

[0055] Color difference calculation: Calculate the colorimetric values ​​synthesized above. Compared with target chromaticity value Substitute into the color difference formula: The overall color difference value is calculated.

[0056] Logical judgment: The system presets a qualified threshold. (NBS unit).

[0057] Scenario 1: If the calculation yields... Since 1.1 < 1.2, the judgment condition is met, the processor sends a "stop printing" command, and the system outputs a "repair qualified" signal.

[0058] Scenario 2: If the calculation yields... Since 4.5 > 1.2, the result is deemed unqualified. The system then feeds back the color difference vector to the iterative formula in claim 3 to calculate the... After adjusting the ink level, continue printing.

[0059] In this embodiment, specifically, the system consists of a high-precision industrial camera (image acquisition unit), an industrial inkjet printer, an industrial control computer (central control processor), and customized image processing software. The software has built-in core algorithm modules, including a defect segmentation module, a color iteration module, and a weight measurement module.

[0060] When the system identifies a dark mineral line defect that needs repair, traditional printing algorithms often result in darker printed colors due to the excessively dark background. This system operates as follows: Step 1: Acquire images of the defective area and extract the background color. The reflectivity of this area was measured. Below the standard value.

[0061] Step 2: Set target color restoration (Usually, the average color of the normal wood surrounding the defect is taken.)

[0062] Step 3: Perform the initial ink volume calculation and print. The industrial camera will immediately capture the printing results after printing. .

[0063] Step 4: Calculate the deviation using an iterative formula. Due to the rough substrate, some ink seeps into the conduit, resulting in... Still dark .

[0064] Step 5: Perform the correction calculation, using the following formula: because Smaller, the correction term in the formula This will produce a significant gain, thereby increasing the amount of ink to compensate for the absorption of the background color.

[0065] Step Six: Perform a second print and check again. If the color difference still exists, continue iterating until... .

[0066] After the repair is completed, it is necessary to determine the degree of integration between the repaired area and the surrounding wood. Since there are natural wood rays and vascular patterns on the surface of the veneer, directly calculating the average color will be affected by the shadows of these patterns.

[0067] For any pixel within the repair area First, calculate its gradient magnitude. .

[0068] If the pixel is located on a texture line and has a large gradient value, it indicates that this is a texture detail rather than the main color difference, according to the formula: The weight of this pixel The smaller size reduces its proportion in the overall color evaluation.

[0069] Conversely, if a pixel is located in a smooth region, the gradient is small and the weight is large.

[0070] Finally, the weighted average formula is used. The obtained color values ​​truly reflect the degree of matching of the base color, rather than the depth of the texture.

[0071] In this embodiment, specifically, taking the processing of a black walnut veneer with natural mineral lines (black stripe defects) as an example, black walnut veneer is a typical non-white substrate (dark base color, low reflectivity) and non-smooth surface (with duct texture). Traditional printing repair solutions have problems such as "grayish after printing, dull color development" and "misjudgment of color difference measurement".

[0072] Step 1: Image Acquisition and Background Color Analysis The system uses an industrial camera with a stable light source to capture images of the surface of the black walnut veneer to be processed.

[0073] The system calculates the background reflectance through weighted analysis. (Relative to a standard whiteboard), and set the reflectivity of the standard whiteboard. Meanwhile, the image processing algorithm identified obvious duct textures in the area, indicating that it is a non-uniform surface.

[0074] Step 2: Establish an iterative printing model for non-white substrates The system extracts the color of the target around the defect. (Lab value: , , ).

[0075] During the initial print run, the system calculates the ink volume and dispenses ink based on a standard algorithm.

[0076] After printing, due to the dark background color of black walnut and its strong ink absorption, the actual color measurement... for( , , The original color was noticeably brighter than the target color (due to insufficient ink coverage, the base color showed through) and lacked sufficient color saturation.

[0077] The system initiates the color iteration correction unit, applying the formula in claim 3: Assuming the color difference vector is at this point .

[0078] Set roughness compensation coefficient (Black walnuts have a rougher surface).

[0079] Calculate the substrate correction factor: The system determines that due to the dark background color, the ink volume correction value needs to be increased by 1.24 times.

[0080] Calculated corrected ink volume This will be 24% more than the usual calculated value to ensure sufficient coverage and saturation.

[0081] Step 3: Closed-loop iterative execution The system performs a second print color correction based on the corrected ink volume, and then captures the image again. The actual color is then measured. near Color difference Once the quality requirements are met, the iteration stops. This process perfectly solves the problem of inaccurate color development on non-white substrates.

[0082] Step 4: Acceptance of Non-Uniform Surface Color Difference Based on Weights After the repair is completed, it is necessary to verify the integration of the repaired area with the surrounding wood. Since the surface of black walnut is covered with dark vascular textures, if the arithmetic average color is calculated directly, the black pixels of the vascular textures will lower the overall average value, causing the system to misjudge the repaired area as "too bright".

[0083] The system activates the weighted color difference measurement unit and applies the formula in claim 5: For each pixel within the repair area, the system calculates its gradient value. .

[0084] Case A (at the conduit texture): Pixel grayscale changes drastically, with large gradients. Assuming... , .

[0085] This pixel has a very low weight in the average calculation, and its color interference is suppressed.

[0086] Case B (flat wood fiber area): uniform pixel grayscale, small gradient, assuming .

[0087] This pixel has a high weight and dominates the final average color calculation.

[0088] The final calculated weighted average color value It primarily reflects the base color of the wood fibers, rather than the vascular bundles. Based on this, the system determines that the repaired area is consistent with the "tone" of the surrounding wood, and thus passes inspection.

[0089] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in the present invention, and these should all be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A digital processing system for natural wood veneer texture images and colors based on machine vision, characterized in that, include: An image acquisition unit is used to acquire anthropomorphic color information features; The color difference recognition and segmentation unit is used to identify the types of natural defects in the electronic original of natural wood veneer, including live knots, dead knots, cracks, mineral lines and color change areas, and to extract the boundary and texture features of the color difference areas; The color iteration correction unit is used to calculate the amount of color correction ink in the defect area for non-white substrates and non-smooth surfaces, and iteratively corrects the color through closed-loop control of printing detection to make the repair result close to the target texture. The weighted color difference measurement unit is used to evaluate the color difference of the repair results of non-uniform surfaces. It calculates the weighted average color value by constructing a weight matrix to avoid the influence of texture on color evaluation. The central control processor is used to coordinate the data flow and logical operations of the above-mentioned units.

2. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 1, characterized in that, The color iteration correction unit performs the following steps: S1. Obtain the base color and establish a color overprinting model for a non-white substrate; S2. Calculate the initial printing ink volume based on the difference between the target color and the base color; S3. After performing the initial color correction, the color information of the printed result is acquired again through the image acquisition unit; S4. Calculate the deviation vector between the current result and the target color, calculate the corrected ink volume change based on the deviation vector, and proceed to the next printing iteration until the color difference value is less than the preset threshold and is jointly confirmed by humans, then terminate the iteration process.

3. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 2, characterized in that, The calculation of establishing the color overprinting model for non-white substrates and correcting the ink volume increment uses the following formula: Let the target color vector be defined. Substrate background color vector , No. The actual color vector for this print is: ; Then the first Ink volume adjustment value required for the next iteration Calculated using the following formula: in, This is the ink volume conversion coefficient matrix. This is the substrate roughness compensation coefficient. The average reflectance of the substrate surface. The reflectance of a standard white substrate; No. The total ink volume for this print run is: .

4. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 1, characterized in that, The weighted color difference measurement unit includes a weight matrix construction module, which assigns weight values ​​to pixels in the image based on the texture frequency features after image capture. Regions with high texture frequency or distinct edge features are assigned lower weight values; Pixels with low texture frequency or in flat areas are assigned higher weight values.

5. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 4, characterized in that, The formula for calculating the weighted average color value is as follows: Let the set of pixels within the repair area be... pixel The chromaticity value is The weight coefficient of this pixel is ; The weighted average color value of the repaired area The calculation formula is: Among them, the weighting coefficient Determined by the following formula: In the formula, For pixels The gradient magnitude at a given point represents the degree of drastic change in texture. This is a texture-sensitive factor used to adjust the degree of influence of non-smooth surfaces on color difference measurements.

6. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 1, characterized in that, The system also includes an ink droplet diffusion simulation module, which is used to simulate the physical diffusion behavior of ink on a non-smooth wood veneer surface before color iteration correction, generate a virtual print preview image, and assist in calculating the initial ink volume.

7. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 1, characterized in that, The types of natural defects in the electronic original of the natural wood veneer include live knots, dead knots, cracks, mineral lines, and discoloration areas. The identification and automatic correction do not change the original characteristics.

8. The digital processing system for natural wood veneer texture images and colors based on machine vision according to claim 1, characterized in that, The color iteration correction unit works in conjunction with the weighted color difference measurement unit, when the weighted average color difference... When the color correction is below the set threshold, the system determines that the color correction is acceptable. After further verification by a human, the iterative printing process stops.

Images