Express carton printing and packaging system and method
Through 3D scanning and image segmentation technology, combined with automatic folding equipment and visual inspection model, the problem of poor printing adaptability of express cartons is solved, efficient and environmentally friendly special-shaped carton printing is achieved, and production efficiency and printing quality are improved.
Patent Information
- Application Number
- CN202510741337.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-05
- Publication Date
- 2025-08-22
- Estimated Expiration
- 2045-06-05
AI Technical Summary
The existing express carton printing and packaging technology is low in efficiency and poor in adaptability, and it is impossible to accurately adapt to the special-shaped carton, resulting in visual distortion of the printing pattern and serious waste of resources, making it difficult to meet the needs of efficiency, environmental protection and personalization.
The three-dimensional data of the paper box is obtained by using 3D scanning technology, combining the phase dewrapping algorithm and the U-Net image segmentation algorithm, accurately calibrate the printable area, optimize the printing pattern layout through the pattern adaptation model, automatically switch the ink type, and combine automatic folding equipment and visual inspection model to achieve automatic production throughout the process.
It realizes high-precision printing adaptation of special-shaped paper boxes, reduces ink consumption, improves production efficiency, reduces visual distortion, ensures printing quality, and supports efficient and environmentally friendly printing with diverse paper boxes shapes.
Smart Images

Figure CN120246379B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of express carton printing, and in particular to an express carton printing and packaging system and method. Background Art
[0002] With the rapid development of e-commerce, the demand for printing and packaging for express cartons, the primary means of transporting goods, has exploded. Traditional printing and packaging technologies for express cartons rely heavily on manual labor or semi-automated equipment, resulting in low efficiency, poor adaptability, and resource waste, making it difficult to meet the growing demand for high efficiency, environmental friendliness, and personalization.
[0003] Existing technologies often rely on fixed templates or simple size adjustments for carton printing, making them incapable of accurately adapting to unusually shaped cartons (such as rounded rectangles and hexagons). This results in visual distortions such as stretching and misalignment of the printed pattern on curved surfaces or at sharp corners. For example, patent CN111814790A discloses a two-dimensional image identification method for automatic identification of aquaculture cages, comprising the following steps: S1: Divide the two-dimensional image space into X×Y segments with equal spacing along the X and Y directions in a rectangular coordinate system, forming X×Y color blocks, and determine the color value range of the color blocks.
[0004] For example, patent CN115723422A discloses a continuous printing method for plastic packaging bags, which relates to the field of plastic packaging bags. The method includes obtaining plastic bag information, including the plastic bag model and the printing position corresponding to the plastic bag model; querying a preset database for a printing pattern corresponding to the plastic bag model, the printing pattern including the printed graphics and printed colors; and generating and executing a pattern update instruction based on the printing pattern corresponding to the plastic bag model. The pattern update instruction is used to update the printed graphics and printed colors. The technology disclosed in this patent lacks the ability to determine defects based on pixel difference thresholds and fails to consider the combined impact of inkjet area and image structure differences, which can easily lead to misjudgments or missed detections. Summary of the Invention
[0005] The present invention provides an express paper box printing and packaging system and method to solve the existing technical problems, and solves the problems in the above-mentioned background technology.
[0006] To solve the above technical problems, according to one aspect of the present invention, more specifically a method for printing and packaging express cartons, the method comprises the following steps:
[0007] S1. Obtain the size, shape parameters and vertex coordinates of the surface printing area of the express paper box based on 3D scanning;
[0008] S2. Real-time connection to the express query interface of the logistics system to generate logistics information including the delivery address, QR code and tracking number;
[0009] S3. Adjust the layout and proportion of the printed pattern based on the size, shape, and vertex coordinates of the printed area on the express box, and optimize the pattern filling using a pattern adaptation model.
[0010] S4. Query the preset express carton recycling grade database and automatically switch between water-based ink and UV curing ink according to the recycling grade;
[0011] S5. After printing is completed, the automatic folding device is triggered to fold the express carton into shape and seal the edges;
[0012] S6. Compare the printed product with the preset standard image based on the visual inspection optimization model, and sort the express cartons with printing defects to the rework line. The visual inspection optimization model determines whether the express cartons have printing defects based on the inkjet area on the express cartons and the difference between the inkjet image and the preset standard image. The results are:
[0013] ;
[0014] Where, Indicates the condition coefficient of printing defects on the printed express paper box. If the printed paper box has a printing defect, it means that it needs to be reprinted. Indicates the difference between the inkjet area on the express paper box after printing and the preset standard image; Indicates the difference between the inkjet image on the express paper box after printing and the preset standard image.
[0015] Furthermore, the specific steps in step S1 are:
[0016] S101, positioning and fully covering 3D scanning of the express carton, and reconstructing the three-dimensional data of the express carton obtained by the scan;
[0017] S102, size and shape parameter extraction;
[0018] S103, calibrating the vertex coordinates of the surface printing area.
[0019] Furthermore, the three-dimensional data of the express box obtained by scanning is reconstructed, and the phase difference of the grating image is calculated based on the phase unwrapping algorithm to generate three-dimensional data of the box surface.
[0020] Furthermore, the size and shape parameter extraction also includes identifying the angular curvature and surface flatness of the paper box through a curvature analysis algorithm, and then determining whether the express paper box is an irregular-shaped box.
[0021] Furthermore, the calibration of the vertex coordinates of the printing area also includes segmenting the printable area on the surface of the paper box using a U-Net image segmentation algorithm, excluding fold lines, seams and existing printed content.
[0022] Furthermore, the specific steps of optimizing pattern filling by the pattern adaptation model are:
[0023] 1) Receive the three-dimensional data of the carton obtained from 3D scanning, including size, shape parameters and vertex coordinates of the printing area;
[0024] 2) Identify high-density filled areas and low-density areas in the pattern based on the U-Net image segmentation algorithm;
[0025] 3) Use dot halftone algorithm to convert continuous tone into discrete dots in low-density areas; use spiral filling inkjet path for high-density filling areas;
[0026] 4) Calculate the minimum ink coverage of each express paper box and set a separate target threshold for each express paper box based on the minimum ink coverage.
[0027] Furthermore, the calculation formula for the minimum ink coverage is:
[0028] ;
[0029] Where, Indicates the minimum ink coverage of each express carton; Indicates that the printable area is divided into After the rectangular grids are formed, the area of the i-th rectangular grid that actually needs to be sprayed with ink; Indicates the total area of the printable area.
[0030] Furthermore, in step S2, the TLS 1.3 protocol can be used to encrypt and transmit logistics information during the real-time connection with the express query interface of the logistics system.
[0031] Furthermore, the formula for calculating the difference between the inkjet image on the printed express paper box and the preset standard image is:
[0032] ;
[0033] In the above formula, Indicates the total number of pixels in the image; Represents the grayscale value of the i-th pixel in the actual printed image; Represents the grayscale value of the i-th pixel in the preset standard image.
[0034] Express carton printing and packaging system, including:
[0035] 3D scanning module, used to capture the three-dimensional data of express cartons in real time;
[0036] Logistics integration module, used to connect to the logistics system API and dynamically generate logistics labels;
[0037] Pattern adaptation model, which optimizes pattern layout based on carton size and calculates minimum ink coverage;
[0038] Ink switching device, used to call different ink storage tanks according to recycling level;
[0039] Automatic folding equipment, integrating pneumatic robotic arms and hot melt glue machines, to achieve automatic forming of express cartons;
[0040] Visual inspection optimization model that uses convolutional neural networks to identify printing defects and sort out defective products.
[0041] Compared with the existing technology, the express paper box printing and packaging system and method provided by the present invention have the following effects:
[0042] 1. The present invention uses 3D scanning technology combined with a phase unwrapping algorithm to obtain accurate three-dimensional data of express cartons in real time, and uses the U-Net image segmentation algorithm to accurately calibrate the printable area, excluding fold lines, seams and existing printed content. Combined with the thin plate spline interpolation algorithm, the pattern is non-rigidly deformed to adapt to the curved paper box, effectively avoiding visual distortion, ensuring that the printed pattern accurately matches the paper box surface, and significantly improving printing quality and adaptability.
[0043] 2. This invention automatically switches between water-based and UV-curable inks based on the carton material and recycling level by querying a recycling grade database. During this switching process, wastewater recycling is employed to reduce pollution. Furthermore, the ink application strategy is optimized through a dot halftoning algorithm and a spiral fill path. Combined with a minimum ink coverage calculation model, the ink usage threshold is dynamically set, significantly reducing ink consumption while balancing environmental performance with cost control.
[0044] 3. The present invention achieves a high degree of automation in the entire process, from real-time generation of logistics information, automatic adaptation of printing patterns, to ink switching, automatic folding and packaging, and visual inspection and sorting. The automatic folding equipment generates the robot arm's motion trajectory in real time according to the carton size and dynamically adjusts the glue injection parameters to ensure folding efficiency and bonding strength. The overall production efficiency is significantly improved compared to traditional manual operations.
[0045] 4. The present invention is based on a visual inspection optimization model, quantifies the printing defect condition coefficient through a difference formula, and automatically sorts unqualified products to the rework line based on a preset threshold, effectively reducing the missed detection rate and lowering the cost of manual re-inspection.
[0046] 5. The present invention uses a curvature analysis algorithm to identify irregular-shaped boxes, and combines a thin plate spline interpolation algorithm and masking technology to ensure that key areas such as QR codes are not deformed, meeting the needs of diversified paper box shapes and high-precision printing. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 is a flow chart of the present invention;
[0048] Figure 2 is a relationship diagram between the condition coefficient g and the difference s in the present invention;
[0049] Figure 3 is a relationship diagram between the condition coefficient g and the difference m in the present invention;
[0050] Figure 4 Graph showing the relationship between the conditional coefficient g and the difference s and difference m in the present invention. DETAILED DESCRIPTION
[0051] In order to make the technical solution of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0052] Example 1, as Figure 1 As shown, according to one aspect of the present invention, a method for printing and packaging express cartons is provided. This method uses 3D scanning to obtain the dimensions and shape parameters of the express cartons, as well as the vertex coordinates of the printed area on the surface. 3D scanning uses technologies such as structured light projection, laser ranging, and a binocular camera to capture the three-dimensional geometric data of the express cartons in real time. This method is used to obtain the dimensions (length, width, height), shape (curvature, angular radius), and vertex coordinates of the printed area on the surface of the cartons. This data also provides basic data for subsequent pattern adaptation, ensuring that the printed pattern accurately matches the carton surface.
[0053] The specific steps to obtain the size, shape parameters and vertex coordinates of the surface printing area of the express paper box based on 3D scanning are:
[0054] 1) The carton enters the scanning area via a conveyor belt. An infrared sensor detects the carton's position and triggers a clamping device to secure it. This step requires a cleaning device (such as an air gun) to remove dust or foreign matter from the carton's surface to ensure scanning accuracy.
[0055] 2) A structured light projector projects a high-density grating pattern onto the carton surface, while a binocular camera array simultaneously captures the deformed grating image. A laser ranging module also assists in obtaining depth information from the carton's edge contours, compensating for data loss caused by structured light on complex curved surfaces.
[0056] 3) Using a phase unwrapping algorithm, the phase difference of the grating image is calculated to generate 3D point cloud data of the carton surface. The laser ranging data is integrated with the structured light point cloud, and the ICP (Iterative Closest Point) algorithm is used to optimize the 3D model accuracy, keeping the error within ±0.1mm.
[0057] The phase unwrapping algorithm is a computational method for processing the phase difference of grating images. It is used to restore the three-dimensional morphology of an object from a deformed grating image. Its function is to convert the phase difference of the grating image acquired by scanning into three-dimensional point cloud data of the carton surface. It can also be combined with laser ranging data and optimize the model accuracy (error ±0.1mm) through the ICP algorithm to ensure the reliability of the scanning results.
[0058] 4) Extract the length, width, and height of the carton from the 3D model. Use a curvature analysis algorithm to identify the carton's corner angles and surface flatness, determining whether it is an irregularly shaped box (e.g., hexagonal, rounded rectangle).
[0059] 5) Use the U-Net image segmentation algorithm to segment the printable area on the carton surface, excluding fold lines, seams, and existing printed content. A coordinate system is established with the lower left corner of the carton as the origin, outputting the vertex and center coordinates of the printable area.
[0060] The U-Net image segmentation algorithm is a deep learning-based image segmentation model that excels at high-precision region segmentation with small sample data. Its purpose is to segment the printable area on the carton surface, excluding fold lines, seams, and existing printed content; and to protect key elements such as logos and QR codes from deformation during pattern adaptation.
[0061] In Example 2, the express query interface of the logistics system is connected in real time to generate logistics information including the delivery address, QR code, and tracking number. This process can also use the TLS 1.3 protocol to encrypt the transmission of logistics information to prevent man-in-the-middle attacks. Furthermore, through a standardized RESTful interface design, it supports seamless integration with the APIs of major logistics systems (such as SF Express and JD.com), and provides parameter mapping templates to accommodate data format differences across different platforms.
[0062] In Example 3, the layout and proportion of the printed pattern are adjusted according to the size, shape, and vertex coordinates of the printed area on the express paper box, and the pattern filling is optimized using a pattern adaptation model. The specific steps of optimizing the pattern filling using the pattern adaptation model are as follows:
[0063] 1) Receive 3D carton data from the 3D scanning module, including dimensions (length, width, height), shape parameters (curvature, corner radius), and vertex coordinates of the printed area. It then loads a preset print pattern (vector or bitmap), analyzes its original resolution, color mode, and key elements (such as logos and text areas), and performs grid segmentation on the printed area to generate a rasterized coordinate matrix covering the printed area.
[0064] 2) Based on the actual carton dimensions and the printed area, the pattern is scaled to the correct size using a bilinear interpolation algorithm, ensuring margins of ≤2mm. Furthermore, if the carton surface is curved (such as a rounded rectangle), a thin plate spline algorithm is used to non-rigidly deform the pattern to compensate for the visual distortion caused by the curve. Masking technology is also used to lock key areas such as the logo and QR code, preventing scaling or deformation to ensure legibility.
[0065] The thin plate spline interpolation algorithm is a non-rigid deformation algorithm used to correct visual distortion when adapting a pattern to a curved surface. It is used to adjust the pattern geometry for rounded corners and irregularly shaped paper box surfaces to avoid stretching or distortion, and combined with the paper box curvature parameters to generate an adaptive printing path in real time.
[0066] 3) Use image segmentation algorithms (such as U-Net) to identify high-density fill areas (such as solid color blocks) and low-density areas (such as gradient backgrounds) within the pattern. A halftone algorithm is used for low-density areas to convert continuous tones into discrete dots, reducing ink usage. For high-density areas, the inkjet path is optimized, using a spiral fill pattern to avoid overlapping sprays.
[0067] The dot halftoning algorithm converts a continuous-tone image into discrete dots, simulating grayscale variations through dot density. This algorithm reduces ink usage by filling low-density areas (such as gradient backgrounds) with dots, thus reducing coverage while maintaining image clarity.
[0068] The spiral fill mode is an inkjet path planning strategy that uses a spiral trajectory to cover high-density fill areas (such as solid color blocks). This method can optimize the inkjet head movement path, reduce ink waste, ensure color consistency in high-density areas, and prevent local over-thickness or missing prints.
[0069] 4) Calculate the minimum ink coverage of each express carton and set a separate target threshold for each express carton based on the minimum ink coverage. The calculation formula for the minimum ink coverage is:
[0070] ;
[0071] Where, Indicates the minimum ink coverage of each express carton; Indicates that the printable area is divided into After the rectangular grids are formed, the area of the i-th rectangular grid that actually needs to be sprayed with ink; Indicates the total area of the printable area.
[0072] Example 4: querying a preset express paper box recycling grade database and automatically switching between water-based ink and UV curing ink according to the recycling grade; the decision logic and pollution control during the switching process are as follows:
[0073] 1) Based on the carton material (corrugated paper / kraft paper), recycling grade (A / B / C), and printing pattern complexity, a decision tree is constructed to select the optimal ink type.
[0074] 2) When switching inks, the flushing program is activated to clean the residual ink in the print head, and the waste water is recycled after sedimentation and filtration.
[0075] Example 5: After printing is completed, the automatic folding device is triggered to fold the express carton into shape and seal the edges. This process also requires robot path optimization and folding quality detection, including:
[0076] 1) Generate the robot arm motion trajectory in real time based on the carton size;
[0077] 2) Dynamically adjust the glue injection speed and temperature according to the ambient temperature and humidity to ensure the bonding strength;
[0078] 3) After folding, the pressure sensor detects the fit of the carton edge. If the pressure value is lower than the threshold (such as 5N), a second fold or an alarm is triggered.
[0079] Example 6: Based on the visual inspection optimization model, the printed product is compared with the preset standard image, and express cartons with printing defects are sorted to the rework line. The visual inspection optimization model determines whether the express cartons have printing defects based on the inkjet area on the express cartons and the difference between the inkjet image and the preset standard image.
[0080] ;
[0081] Where, Indicates the condition coefficient of printing defects on the printed express paper box. If the printed paper box has a printing defect, it means that it needs to be reprinted. Indicates the difference between the inkjet area on the express paper box after printing and the preset standard image; Indicates the difference between the inkjet image on the express paper box after printing and the preset standard image.
[0082] The above formula is an analysis conducted to solve practical problems in enterprises, that is, empirical analysis. The data obtained in turn and the characteristic relationships between the data are the external manifestations of the empirical formula. The reasoning process is as follows:
[0083] 1) Fit the formula of the conditional coefficient g for the presence of printing defects on printed express cartons.
[0084] Among them, the conditional coefficient g of the presence of printing defects in the printed express paper box can be expressed by the number of defects in the sample express paper box.
[0085] For example, data from 100 express paper box samples are collected. If the number of defects in a certain express paper box exceeds the data in the other 50 samples after manual evaluation, then it means that the conditional coefficient g=50% that the printed express paper box has printing defects.
[0086] Other data are calculated and fitted using data collected by the device itself.
[0087] 2) Establish a mathematical model for the relationship between the conditional coefficient g and the difference s (such as Figure 2 As shown in the figure, the red dots are the distribution of the 100 samples collected), then:
[0088] (Formula 1)
[0089] In the above formula 1, k represents an empirical constant for adjusting the sensitivity of the above model.
[0090] 3) Establish a mathematical model for the relationship between the conditional coefficient g and the difference m (such as Figure 3 As shown in the figure, the blue dots are the distribution of the 100 samples collected), then:
[0091] (Formula 2)
[0092] In the above formula 2, k represents an empirical constant for adjusting the sensitivity of the above model.
[0093] 4) Establish a mathematical model for the relationship between the conditional coefficient g and the difference s and difference m (such as Figure 4 As shown, the model shows that the difference s and the difference m are positively correlated), and combined with the characteristic relationship of the above formula 1 and formula 2, we have:
[0094] g=(Formula 1)×(Formula 2)
[0095] According to the above derivation, we can know that the conditional coefficient g for the presence of printing defects in the printed express paper box is:
[0096] ;
[0097] The calculation formula for the difference between the inkjet area on the express paper box after printing and the preset standard image is:
[0098] ;
[0099] In the above formula, Indicates the actual inkjet area; Indicates the inkjet area of a preset standard image.
[0100] The calculation formula for the difference between the inkjet image on the printed express paper box and the preset standard image is:
[0101] ;
[0102] In the above formula, Indicates the total number of pixels in the image; Represents the grayscale value of the i-th pixel in the actual printed image; Represents the grayscale value of the i-th pixel in the preset standard image.
[0103] Then, when data sampling and quality inspection are performed on a certain express paper box, the difference between the inkjet area on the express paper box after printing and the preset standard image is obtained. ; The difference between the inkjet image on the express paper box after printing and the preset standard image is measured , then we have:
[0104]
[0105] According to the above calculation, we can know that the condition coefficient of printing defects on the printed express paper box is , then it means that there are printing defects on the printed express paper box, and it needs to be reprinted. And compare multiple sets of data:
[0106] Table 1 Partial implementation data and actual printing status of express cartons
[0107]
[0108] Based on the data in Table 1 above, we can know that when the sample data tends to infinity, the condition coefficient g value will be used as a dividing line to divide the printing status of the express paper box. As can be seen from the data in Table 1 above, this dividing line is 68.2%. When manual evaluation shows that the number and size of defects are obvious, it means that there are printing defects on the printed express paper box and it needs to be reprinted.
[0109] Example 7, express carton printing and packaging system, comprising:
[0110] 3D scanning module, used to capture the three-dimensional data of express cartons in real time;
[0111] Logistics integration module, used to connect to the logistics system API and dynamically generate logistics labels;
[0112] Pattern adaptation model, which optimizes pattern layout based on carton size and calculates minimum ink coverage;
[0113] Ink switching device, used to call different ink storage tanks according to recycling level;
[0114] Automatic folding equipment, integrating pneumatic robotic arms and hot melt glue machines, to achieve automatic forming of express cartons;
[0115] The visual inspection optimization model uses convolutional neural networks to identify printing defects and sort out defective products. Convolutional neural networks are deep learning models that excel at extracting and classifying local image features. They can be used to identify blur, jagged edges, or color deviations in printed products, routing defective products to rework lines and reducing manual intervention.
[0116] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.
Claims
1. Express carton printing and packaging method, characterized in that: The following steps are involved: S1. Obtain the size, shape parameters and vertex coordinates of the surface printing area of the express paper box based on 3D scanning; S2. Real-time connection to the express query interface of the logistics system to generate logistics information including the delivery address, QR code and tracking number; S3. Adjust the layout and proportion of the printed pattern based on the size, shape, and vertex coordinates of the printed area on the express box, and optimize the pattern filling using a pattern adaptation model. Based on the actual size of the carton and the printing area, the pattern is scaled to an appropriate size using a bilinear interpolation algorithm. If the carton surface is curved, a thin plate spline interpolation algorithm is used to perform non-rigid deformation on the pattern to compensate for the visual distortion caused by the curved surface. S4. Query the preset express carton recycling grade database and automatically switch between water-based ink and UV curing ink according to the recycling grade; S5. After printing is completed, the automatic folding device is triggered to fold the express carton into shape and seal the edges; S6. Compare the printed product with the preset standard image based on the visual inspection optimization model, and sort the express cartons with printing defects to the rework line. The visual inspection optimization model determines whether the express cartons have printing defects based on the inkjet area on the express cartons and the difference between the inkjet image and the preset standard image. The calculation formula is: ; Where, Indicates the condition coefficient of printing defects on the printed express paper box. If the printed paper box has a printing defect, it means that it needs to be reprinted. Indicates the difference between the inkjet area on the express paper box after printing and the preset standard image; Indicates the difference between the inkjet image on the express paper box after printing and the preset standard image; The specific steps of optimizing pattern filling by the pattern adaptation model are as follows: 1) Receive the three-dimensional data of the carton obtained from 3D scanning, including size, shape parameters and vertex coordinates of the printing area; 2) Identify high-density filled areas and low-density areas in the pattern based on the U-Net image segmentation algorithm; 3) Use dot halftone algorithm to convert continuous tone into discrete dots in low-density areas; use spiral filling inkjet path for high-density filling areas; 4) Calculate the minimum ink coverage of each express paper box and set a separate target threshold for each express paper box based on the minimum ink coverage.
2. The express paper box printing and packaging method according to claim 1, characterized in that: The specific steps in step S1 are: S101, positioning and fully covering 3D scanning of the express carton, and reconstructing the three-dimensional data of the express carton obtained by the scan; S102, size and shape parameter extraction; S103, calibrating the vertex coordinates of the surface printing area.
3. The express paper box printing and packaging method according to claim 2, characterized in that: The three-dimensional data of the express paper box obtained by scanning is reconstructed, and the phase difference of the grating image is calculated based on the phase unwrapping algorithm to generate three-dimensional data of the paper box surface.
4. The express paper box printing and packaging method according to claim 2, characterized in that: The size and shape parameter extraction also includes identifying the angular curvature and surface flatness of the paper box through a curvature analysis algorithm, and then determining whether the express paper box is an irregular-shaped box.
5. The express paper box printing and packaging method according to claim 2, characterized in that: The calibration of the vertex coordinates of the printing area also includes segmenting the printable area on the surface of the paper box using the U-Net image segmentation algorithm, excluding fold lines, seams and existing printed content.
6. The express paper box printing and packaging method according to claim 1, characterized in that: The calculation formula for the minimum ink coverage is: ; Where, Indicates the minimum ink coverage of each express carton; Indicates that the printable area is divided into After the rectangular grids are formed, the area of the i-th rectangular grid that actually needs to be sprayed with ink; Indicates the total area of the printable area.
7. The express paper box printing and packaging method according to claim 1, characterized in that: In step S2, the TLS 1.3 protocol can be used to encrypt and transmit logistics information during the real-time connection to the express query interface of the logistics system.
8. The express paper box printing and packaging method according to claim 1, characterized in that: The calculation formula for the difference between the inkjet image on the printed express paper box and the preset standard image is: ; In the above formula, Indicates the total number of pixels in the image; Represents the grayscale value of the i-th pixel in the actual printed image; Represents the grayscale value of the i-th pixel in the preset standard image.
9. Express carton printing and packaging system, characterized by: The express carton printing and packaging method according to any one of claims 1 to 8, wherein the express carton printing and packaging system comprises: 3D scanning module, used to capture the three-dimensional data of express cartons in real time; Logistics integration module, used to connect to the logistics system API and dynamically generate logistics labels; Pattern adaptation model, which optimizes pattern layout based on carton size and calculates minimum ink coverage; Ink switching device, used to call different ink storage tanks according to recycling level; Automatic folding equipment, integrating pneumatic robotic arms and hot melt glue machines, to achieve automatic forming of express cartons; Visual inspection optimization model that uses convolutional neural networks to identify printing defects and sort out defective products.
Citation Information
Patent Citations
Two-dimensional image identification method for automatic identification of aquaculture net cage
CN111814790A
Miniature laser three-dimensional model reconstruction system and method
CN113624159A
Printing defect vision detection system
CN206114546U
Method for designing two-dimensional graphics for use on three-dimensional cartons bearing such graphics
US20050157342A1