Gold stamping packaging box production method and system

Abstract

The invention relates to a gold stamping packaging box production method and system, and the method comprises the following steps: S1, carrying out the humidity adjustment and flatness correction of a packaging box blank raw material, and enabling the humidity and flatness of the raw material to reach preset threshold values; s2, a gold stamping target area of the blank is recognized, and three-dimensional coordinates of the gold stamping target area are obtained; s3, based on the raw material characteristics, the environmental parameters and the positioning data, gold stamping temperature, pressure and speed parameters are calculated through a machine learning model; s4, according to the three-dimensional coordinates and the gold stamping parameters, a gold stamping plate is controlled to be attached to the green body for hot stamping; s5, performing defect detection on the gold-stamped green body, and dividing qualified products and unqualified products according to a detection result; and S6, the qualified product is subjected to cutting, film covering, impurity removal and collection. The method has the beneficial effects that substrate deformation caused by non-uniform humidity of raw materials and poor flatness is eliminated, the three-dimensional space coordinates of the gold stamping target area can be accurately recognized, and the gold stamping temperature, pressure and speed can be dynamically optimized based on a machine learning model.

Description

Technical Field

[0001] This invention relates to the field of hot stamping product technology, specifically to a method and system for producing hot stamping packaging boxes. Background Technology

[0002] Gold foil stamping is a key surface treatment technology for enhancing the appearance and texture of packaging boxes and increasing product added value. It is widely used in the production of packaging boxes for high-end gifts, cosmetics, tobacco, and alcohol products. Traditional gold foil stamping production mainly relies on manual experience or semi-automated equipment, which has many limitations.

[0003] In the existing technology, the raw materials of packaging box blanks are easily affected by the temperature and humidity of the environment. If hot stamping is carried out directly, it will lead to defects such as inaccurate hot stamping position and incomplete hot stamping. In addition, the setting of hot stamping process parameters depends heavily on the experience of the operators and lacks the ability to adapt to environmental changes and other factors. The quality of products varies greatly between different batches or even within the same batch. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application proposes a hot stamping packaging box production method and system. Through raw material pretreatment, it effectively eliminates substrate deformation caused by uneven raw material humidity and poor flatness. It can accurately identify the three-dimensional spatial coordinates of the hot stamping target area and dynamically optimize the hot stamping temperature, pressure, and speed based on a machine learning model. This ensures that the process parameters are always matched with the raw material characteristics and environmental conditions, thereby ensuring the optimization of the hot stamping effect and batch-to-batch stability.

[0005] The following is the technical solution of the present invention: a method for producing hot stamping packaging boxes, comprising the following steps: S1. Adjust the humidity and flatness of the raw material of the packaging box blank so that the humidity and flatness of the raw material reach the preset threshold. S2. Identify the target area for hot stamping on the blank and obtain its three-dimensional coordinates; S3. Based on raw material characteristics, environmental parameters, and positioning data, calculate hot stamping temperature, pressure, and speed parameters using a machine learning model; S4. Based on the three-dimensional coordinates and hot stamping parameters, control the hot stamping plate to adhere to the blank for hot stamping. S5. Perform defect detection on the hot-stamped blanks and sort qualified and unqualified products according to the detection results; S6. Cut, coat, remove impurities and collect qualified products.

[0006] As a preferred embodiment of the present invention, S1 includes the following steps: S101. Transfer the billet raw material to the processing area; S102. Collect the moisture content and flatness error value of the raw materials; S103. Compare the humidity value with the preset threshold, and start the humidifier or dehumidifier to adjust the humidity; S104. Adjust the pressure of the pressure rollers according to the flatness error value to correct the flatness. S105. Continuously monitor the data. If the standard is met, proceed to S2; otherwise, reprocess.

[0007] As a preferred embodiment of the present invention, S2 includes the following steps: S201, preheating industrial camera and laser rangefinder sensor; S202. Simultaneously capture images of the billet and collect height data; S203. Perform grayscale conversion, filtering, and edge detection processing on the image; S204. Match feature points based on the SIFT algorithm and calculate 3D coordinates by combining height data.

[0008] As a preferred embodiment of the present invention, S3 includes the following steps: S301. Collect data on raw material moisture content, flatness error, ambient temperature, ambient humidity, and positioning height, as well as historical production data; S302. Normalize the data; S303. Use the random forest regression algorithm to train the parameter prediction model, input real-time data into the model, and predict the hot stamping parameters. S304. Compare the predicted parameters with the safety threshold and output the final parameters.

[0009] As a preferred embodiment of the present invention, S4 includes the following steps: S401: Receive positioning coordinates and hot stamping parameters, adjust the hot stamping plate to the initial position and start heating; S402: Temperature is controlled by a PID temperature control algorithm, and the position of the hot stamping plate is adjusted according to the coordinates; S403. Adjust the pressure to the set value and control the hot stamping plate to descend at the set speed for hot stamping; S404. After hot stamping is completed, reset the hot stamping plate and transfer the blank.

[0010] As a preferred embodiment of the present invention, S5 includes the following steps: S501, Preheat the sensor and set the defect judgment threshold; S502. Collect the spectral image, RGB image, and gloss value of the hot stamping area; S503, Extract color, spectral and shape features; S504. Use a CNN model to classify defects and combine it with gloss to determine acceptance. S505. Based on the judgment results, product diversion shall be carried out.

[0011] As a preferred embodiment of the present invention, S6 includes the following steps: S601. Laser cut the blank according to the preset dimensions; S602. Apply heat-pressing film to the cut product; S603. Remove surface impurities using an ultrasonic cleaning device; S604. Count and pack the products.

[0012] In a preferred embodiment of the present invention, in S403, the hot stamping time is calculated based on pressure, temperature, and speed. The expression is as follows: , In the above formula, For hot stamping time, Based on the hot stamping time, For the synergy coefficient, For correction factor, This refers to the hot stamping temperature. For hot stamping pressure, This refers to the speed of hot stamping.

[0013] As a preferred embodiment of the present invention, in S601, the expression for calculating the cutting path coordinates is as follows: , In the above formula, These are the vertex coordinates of the clipping path. The center coordinates of the packaging box blank These are the length and width of the packaging box, respectively.

[0014] A hot stamping packaging box production system, comprising: The raw material pretreatment module is used to adjust the humidity and correct the flatness of the packaging box blank raw material; The fusion positioning module is used to identify the hot stamping target area of ​​the packaging box blank, obtain its coordinate information in three-dimensional space, and connect to the raw material pretreatment module; The hot stamping parameter adjustment module calculates the appropriate hot stamping temperature, pressure, and speed parameters based on raw material characteristics, environmental parameters, positioning data, and historical production data, and connects to the fusion positioning module. The hot stamping execution module is used to complete the bonding, hot stamping, and separation of the hot stamping plate and the packaging box blank according to the positioning data and parameters. It connects the hot stamping parameter adjustment module and the fusion positioning module. The quality inspection module is used to inspect the blanks after hot stamping, identify defect types, and sort the products. It is connected to the hot stamping execution module. The finished product post-processing module cuts, coats, and removes impurities from qualified hot stamping blanks to obtain finished hot stamping packaging boxes, which are then connected to the quality inspection module.

[0015] The beneficial effects of this invention are: 1. In this invention, the pretreatment of raw materials effectively eliminates the substrate deformation caused by uneven moisture content and poor flatness of the raw materials, providing a stable and qualified blank for subsequent high-precision hot stamping; 2. In this invention, the sub-millimeter level three-dimensional spatial coordinates of the hot stamping target area are accurately identified through fusion positioning, overcoming the shortcomings of traditional methods that cannot compensate for individual deformations, and improving the positional accuracy and consistency of the hot stamping pattern. 3. In this invention, the hot stamping parameters are adaptively adjusted, and the hot stamping temperature, pressure, and speed are dynamically optimized based on a machine learning model. This ensures that the process parameters always match the characteristics of the raw materials and environmental conditions, avoiding hot stamping problems caused by improper parameters and ensuring the optimization of the hot stamping effect and batch-to-batch stability. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system of the present invention; Figure 2 This is a diagram illustrating the steps of the method of the present invention; Figure 3 This is a flowchart of the method of the present invention; In the diagram: 1. Raw material pretreatment module; 2. Fusion positioning module; 3. Hot stamping parameter adjustment module; 4. Hot stamping execution module; 5. Quality inspection module; 6. Finished product post-processing module. Detailed Implementation

[0017] To make the technical problems solved by the present invention, the technical solutions adopted, and the technical effects achieved clearer, the technical solutions of the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 like Figure 1 As shown, a hot stamping packaging box production system includes: Raw material pretreatment module 1 is used to adjust the humidity and correct the flatness of the packaging box blank raw material; The fusion positioning module 2 is used to identify the hot stamping target area of ​​the packaging box blank, obtain its coordinate information in three-dimensional space, and connect to the raw material pretreatment module 1; The hot stamping parameter adjustment module 3 calculates the appropriate hot stamping temperature, pressure, and speed parameters based on raw material characteristics, environmental parameters, positioning data, and historical production data, and connects to the fusion positioning module 2. Hot stamping execution module 4 is used to complete the bonding, hot stamping and separation of hot stamping plate and packaging box blank according to positioning data and parameters, and connects hot stamping parameter adjustment module 3 and fusion positioning module 2; Quality inspection module 5 is used to inspect the blank after hot stamping, identify the defect type and sort the products, and is connected to hot stamping execution module 4; The finished product post-processing module 6 cuts, coats, and removes impurities from the qualified hot stamping blank to obtain the finished hot stamping packaging box, which is then connected to the quality inspection module 5.

[0019] In this embodiment, the raw material pretreatment module 1 is used to regulate the humidity and correct the flatness of the packaging box blank raw material, eliminating deformation caused by environmental changes during raw material storage. The raw material pretreatment module 1 includes a humidity regulation unit, a flatness correction unit, a raw material conveying unit, and a status monitoring unit. The humidity regulation unit is equipped with an ultrasonic humidifier and a dehumidifier. The flatness correction unit adopts a double-roller pressing structure, with the rollers covered in silicone material. The input end of the raw material pretreatment module 1 is connected to the raw material storage outlet via a conveyor belt, and the output end is connected to the input end of the fusion positioning module 2 via a servo-driven conveyor belt. The status monitoring unit transmits raw material humidity and flatness data to the PLC main controller via an RS485 bus. The PLC main controller sends control commands to the humidity regulation unit and the flatness correction unit.

[0020] After the raw materials enter the module, the status monitoring unit first collects the humidity value. Flatness error value The PLC main controller compares the collected data with the preset threshold. If the humidity value is greater than the humidity threshold, the dehumidifier is started. If the humidity value is less than the humidity threshold, the humidifier is started. At the same time, the flatness correction unit adjusts the double roller pressing pressure according to the flatness error value until the raw material condition meets the standard, and then it is conveyed to the fusion positioning module 2 by the conveying unit.

[0021] In this embodiment, the fusion positioning module 2 is used to identify the hot stamping target area of ​​the packaging box blank and obtain its coordinate information in three-dimensional space, providing a positioning reference for the hot stamping execution module 4. The fusion positioning module 2 includes a left industrial camera, a right industrial camera, a laser rangefinder, an image acquisition card, and a coordinate calculation unit. The industrial cameras are installed above the raw material conveying path at a distance of 500mm, and the laser rangefinder is installed below and between the two cameras, pointing vertically towards the surface of the raw material. The input end of the fusion positioning module 2 receives the raw material conveyed by the raw material preprocessing module 1. The industrial cameras and the laser rangefinder are connected to the coordinate calculation unit through the image acquisition card. The coordinate calculation unit transmits the three-dimensional coordinate data to the PLC main controller via industrial Ethernet. The PLC main controller synchronizes the positioning data to the hot stamping parameter adjustment module 3 and the hot stamping execution module 4.

[0022] After the raw material enters the positioning area, the left and right industrial cameras capture images of the billet, while a laser rangefinder collects surface height data. The image acquisition card converts the image data into digital signals, and the coordinate calculation unit performs grayscale conversion and edge detection on the image to extract feature points of the hot stamping target area. Combining the laser rangefinder data with a coordinate fusion algorithm, the three-dimensional coordinates of the target area are calculated. The coordinate data is transmitted to the PLC main controller for caching.

[0023] In this embodiment, the hot stamping parameter adjustment module 3 calculates suitable hot stamping temperature, pressure, and speed parameters based on raw material characteristics, environmental parameters, positioning data, and historical production data. The hot stamping parameter adjustment module 3 includes a data storage unit, a feature extraction unit, a model training unit, and a parameter output unit. The hot stamping parameter adjustment module 3 receives the raw material moisture value transmitted from the PLC main controller via industrial Ethernet. Flatness error value Positioning coordinates Ambient temperature Ambient humidity The data and parameter output unit transmits the hot stamping parameters via the Profinet bus. The data is transmitted to the hot stamping execution module 4, and the parameter data is simultaneously sent back to the PLC main controller for storage.

[0024] First, the historical production data sample set of qualified products is retrieved from the data storage unit; the feature extraction unit normalizes the input data and sample data, and extracts... , , , and Feature vector; the model training unit trains the sample data based on the random forest regression algorithm to generate a parameter prediction model; the parameter output unit inputs the real-time feature vector into the model and outputs the hot stamping temperature. , Pressure , Speed Optimal values.

[0025] In this embodiment, the hot stamping execution module 4 is used to complete the fitting, hot stamping, and separation of the hot stamping plate and the blank of the packaging box according to the positioning data and parameters. The hot stamping execution module 4 includes a hot stamping plate installation unit, a servo drive unit, a pressure adjustment unit, a temperature control unit, and a speed adjustment unit. Among them, the hot stamping plate is made of copper material and its surface is chrome-plated. The servo drive unit uses a linear motor, the pressure adjustment unit uses a pneumatic proportional valve, and the temperature control unit uses a PID temperature controller. The input end of the hot stamping execution module 4 receives the positioned raw materials through a conveyor belt and receives the parameters of the hot stamping parameter adjustment module 3 and the positioning data of the PLC main controller through the Profinet bus; the actual temperature, pressure, speed and other status data of the module are transmitted back to the PLC main controller through the industrial Ethernet.

[0026] After the raw materials reach the hot stamping station, the servo drive unit adjusts the horizontal position of the hot stamping plate according to the positioning coordinates and adjusts the height of the hot stamping plate according to the coordinates; the temperature control unit heats the hot stamping plate to the preset temperature , and the pressure adjustment unit adjusts the output pressure to ; the servo drive unit drives the hot stamping plate to descend at a speed of and keeps the hot stamping time when it is in contact with the blank, which is 0.5 to 2 seconds; after the hot stamping is completed, the hot stamping plate quickly rises and resets, and the conveyor belt conveys the blank to the quality inspection module 5.

[0027] In this embodiment, the quality inspection module 5 is used to detect the hot-stamped blanks, identify four types of defects including hot stamping incompleteness, position misalignment, color difference deviation, and unqualified glossiness, and classify and label the defect types and severity levels. The quality inspection module 5 includes a hyperspectral camera, a glossiness sensor, an image processing unit, and a defect classification unit. Among them, the hyperspectral camera and the glossiness sensor are installed above the inspection station with a spacing of 300 mm and are vertically directed at the surface of the blank. The input end of the quality inspection module 5 is connected to the output end of the hot stamping execution module 4 through a conveyor belt, and the hyperspectral camera and the glossiness sensor transmit the collected data to the image processing unit; the defect classification unit transmits the inspection results such as whether it is qualified, defect type, and defect level to the PLC main controller through the industrial Ethernet.

[0028] After the hot-stamped blank enters the detection area, the hyperspectral camera collects the spectral image and RGB image of the hot stamping area, and the glossiness sensor collects the glossiness value The image processing unit performs noise reduction and enhancement on the image, extracting the color, spectral, and shape features of the hot stamping area; the defect classification unit analyzes the feature data based on a CNN model, combining it with gloss values. The system determines whether a product is qualified and what type of defect it is. The CNN model structure includes an input layer, three convolutional layers, two pooling layers, a fully connected layer, and an output layer. After the inspection is completed, qualified products are sent to the finished product post-processing module 6, while unqualified products are marked and diverted to the waste recycling unit.

[0029] In this embodiment, the finished product post-processing module 6 cuts and shapes the qualified hot stamping blanks, applies a protective film to the surface, and removes surface impurities to obtain the finished hot stamping packaging box. The finished product post-processing module 6 includes a cutting unit, a film-coating unit, a surface finishing unit, and a finished product collection unit. The cutting unit uses a laser cutting machine, the film-coating unit uses a hot-press film-coating structure, and the surface finishing unit uses an ultrasonic impurity removal device. The input end of the finished product post-processing module 6 is connected to the qualified product output end of the quality inspection module 5 via a conveyor belt. It receives control commands through the PLC main controller, and its operating status data is transmitted back to the PLC main controller in real time.

[0030] After the qualified blanks enter the finished product post-processing module 6, the cutting unit cuts them according to the preset packaging box size parameters; the cut semi-finished products are transferred to the laminating unit, where a transparent protective film is covered by hot pressing; the laminated products enter the appearance finishing unit, where an ultrasonic impurity removal device removes surface dust and residual impurities; finally, the finished product collection unit sorts, counts, and packs them.

[0031] Example 2 like Figure 2 and Figure 3 As shown, a method for producing hot stamping packaging boxes includes the following steps: S1. Adjust the humidity and flatness of the raw material of the packaging box blank so that the humidity and flatness of the raw material reach the preset threshold. S2. Identify the target area for hot stamping on the blank and obtain its three-dimensional coordinates; S3. Based on raw material characteristics, environmental parameters, and positioning data, calculate hot stamping temperature, pressure, and speed parameters using a machine learning model; S4. Based on the three-dimensional coordinates and hot stamping parameters, control the hot stamping plate to adhere to the blank for hot stamping. S5. Perform defect detection on the hot-stamped blanks and sort qualified and unqualified products according to the detection results; S6. Cut, coat, remove impurities and collect qualified products.

[0032] In step S1, raw material pretreatment is performed, including humidity adjustment and flatness correction of the packaging box blank raw material to bring the raw material humidity and flatness to a preset threshold. This includes the following steps: S101. Transfer the billet raw material to the processing area; Start the conveying unit of raw material pretreatment module 1 to transfer the packaging box blank raw materials in the warehouse to the processing area. The conveying speed is set to 1m / min.

[0033] S102. Collect the moisture content and flatness error value of the raw materials; Humidity sensor collects raw material humidity value The flatness detection sensor uses laser ranging to collect the height difference of various points on the surface of the raw material and calculates the flatness error. The sampling frequency is 10Hz.

[0034] S103. Compare the humidity value with the preset threshold, and start the humidifier or dehumidifier to adjust the humidity; The PLC main controller will display the humidity value. and humidity threshold In comparison, if Turn on the dehumidifier; if Turn on the ultrasonic humidifier to stabilize the humidity within the range of 5%-8%.

[0035] S104. Adjust the pressure of the pressure rollers according to the flatness error value to correct the flatness. Based on flatness error Control the double-roller pressing pressure of the flatness correction unit. Increase the pressure of the pressure roller. Reduce pressure and keep the speed of the pressure roller and the conveying speed synchronized to ensure that the flatness error of the raw material is ≤0.1mm.

[0036] S105. Continuously monitor the data. If the standard is met, transmit it to step S2; otherwise, reprocess it. The status monitoring unit continuously collects three sets of humidity and flatness data. If all of them meet the threshold requirements, the raw material pretreatment is deemed qualified, and the conveying unit transfers the raw material to the fusion positioning module 2. If it is not qualified, it returns to sub-step S103 for reprocessing.

[0037] In step S2, pre-hot stamping positioning calibration is performed. The hot stamping target area of ​​the blank is identified by a dual-vision-laser fusion positioning system, and its three-dimensional coordinates are obtained. This includes the following steps: S201, preheating industrial camera and laser rangefinder sensor; The industrial camera and laser rangefinder sensor warm up for 30 seconds, the image acquisition card is initialized, and the coordinate calculation unit enters the ready state.

[0038] S202. Simultaneously capture images of the billet and collect height data; After the raw material enters the positioning area, the PLC main controller sends a trigger signal, and the left and right industrial cameras simultaneously capture images of the raw material surface. The laser rangefinder sensor collects the height data of the center position of the raw material surface. .

[0039] S203. Perform grayscale conversion, filtering, and edge detection processing on the image; The acquired images are processed by grayscale conversion, Gaussian filtering for noise reduction, and Canny edge detection to extract the contour feature points of the hot stamping target area.

[0040] S204. Match feature points based on the SIFT algorithm and calculate 3D coordinates by combining height data; Pixel coordinates are obtained by matching feature points in two camera images using the SIFT algorithm. and Combined with laser ranging data The three-dimensional coordinates of the target area are calculated using a coordinate fusion algorithm. .

[0041] The 3D coordinates are calculated using SIFT feature extraction and coordinate fusion, as shown in the following expression: , In the above formula, The pixel coordinates of the feature points acquired by the left industrial camera. The pixel coordinates of the feature points acquired by the right industrial camera. The focal length of the left industrial camera. The focal length of the industrial camera on the right. The distance between the left and right industrial cameras. The height data collected by the laser rangefinder sensor. This is the reference height.

[0042] In step S3, the hot stamping parameters are adjusted. Based on the characteristics of the raw materials, environmental parameters, and positioning data, the appropriate hot stamping temperature, pressure, and speed parameters are calculated using a machine learning model, including the following steps: S301. Collect data on raw material moisture content, flatness error, ambient temperature, ambient humidity, and positioning height, as well as historical production data; Receive raw material moisture value transmitted from PLC main controller Flatness error value Ambient temperature Ambient humidity Positioning height data And historical production data for nearly 1,000 qualified products.

[0043] S302. Normalize the data; The collected data is normalized, mapping each parameter to the [0,1] interval. The normalization formula is as follows: , In the above formula, This is the minimum value of the parameter. For the maximum value of the parameter, For parameter values.

[0044] S303. Use the random forest regression algorithm to train the parameter prediction model, input real-time data into the model, and predict the hot stamping parameters. Based on the random forest regression algorithm, using the preprocessed feature vector as input and the hot stamping temperature of historical qualified products... ,pressure ,speed For the output, 120 decision trees are trained to form a parameter prediction model.

[0045] Input the real-time feature vector into the trained model, and output the predicted hot stamping temperature, pressure, and speed.

[0046] S304. Compare the predicted parameters with the safety threshold and output the final parameters; The predicted parameters are compared with the safety threshold range. If they are within the range, they are output directly; if they are outside the range, the threshold boundary value is taken as the final parameter.

[0047] In step S4, high-precision hot stamping is performed. Based on the three-dimensional coordinates and hot stamping parameters, the hot stamping plate is controlled to adhere to the blank for hot stamping, including the following steps: S401: Receive positioning coordinates and hot stamping parameters, adjust the hot stamping plate to the initial position and start heating; Hot stamping execution module 4 receives positioning coordinates and hot stamping temperature ,pressure ,speed Once the hot stamping parameters are set, the servo drive unit adjusts the hot stamping plate to its initial position, and the temperature control unit begins heating the hot stamping plate.

[0048] S402: Temperature is controlled by a PID temperature control algorithm, and the position of the hot stamping plate is adjusted according to the coordinates; A PID temperature control algorithm is used, based on the actual temperature of the hot stamping plate and the target temperature. The heating power is adjusted according to the deviation. The servo drive unit adjusts the heating power based on the positioning coordinates. Adjust the horizontal position of the hot stamping plate; according to Adjust the initial height of the hot stamping plate to make the initial distance between the hot stamping plate and the surface of the raw material 5mm.

[0049] S403. Adjust the pressure to the set value and control the hot stamping plate to descend at the set speed for hot stamping; The pressure regulating unit adjusts the output pressure to The servo drive unit operates according to speed. The hot stamping plate is driven down to adhere to the surface of the raw material for hot stamping.

[0050] The hot stamping time is calculated based on pressure, temperature, and speed. The expression is as follows: , In the above formula, For hot stamping time, Based on the hot stamping time, For the synergy coefficient, For correction factor, This refers to the hot stamping temperature. For hot stamping pressure, This refers to the speed of hot stamping.

[0051] S404. After hot stamping is completed, reset the hot stamping plate and transfer the blank. After hot stamping is completed, the servo drive unit drives the hot stamping plate to rise and reset at a speed of 5m / min, and the conveyor belt transports the hot stamped material to the quality inspection module 5.

[0052] In step S5, quality inspection is performed, and defects are detected on the hot-stamped blanks. Based on the inspection results, qualified and unqualified products are separated, including the following steps: S501, Preheat the sensor and set the defect judgment threshold; Quality inspection module 5 starts, hyperspectral camera and gloss sensor warm up for 20 seconds, image processing unit and defect classification unit are initialized, and defect judgment threshold is set.

[0053] S502. Collect the spectral image, RGB image, and gloss value of the hot stamping area; After hot stamping, the raw material enters the inspection area. A hyperspectral camera acquires the spectral and RGB images of the hot stamping area, while a gloss sensor collects gloss values ​​at three different locations. , , Take the average gloss level .

[0054] S503, Extract color, spectral and shape features; Color space conversion is performed on RGB images to extract hue, saturation, and brightness features; principal component analysis is performed on spectral images to extract the first three principal components as spectral features; shape features such as area, perimeter, and roundness are extracted from contour images, and feature vectors are extracted.

[0055] S504. Use a CNN model to classify defects and combine it with gloss to determine acceptance. For defect classification, the feature vector is input into the CNN defect classification model, and the model outputs the probability value of each type of defect. If the category corresponding to the highest probability is qualified and the average gloss is greater than or equal to the defect judgment threshold, the product is judged to be qualified; otherwise, it is judged to be unqualified and the defect type is marked.

[0056] S505. Based on the judgment results, product diversion shall be carried out; The products are then diverted. The PLC main controller controls the diversion device based on the detection results. Qualified products are sent to the finished product post-processing module 6, while unqualified products are sent to the waste recycling unit.

[0057] In step S6, post-processing of finished products is carried out, including cutting, laminating, removing impurities, and collecting qualified products, which includes the following steps: S601. Laser cut the blank according to the preset dimensions; The cutting unit of the finished product post-processing module 6 receives qualified products and cuts the product outline according to the preset packaging box size parameters using a laser cutting machine. The cutting speed is synchronized with the conveying speed.

[0058] The expression for calculating the cropping path coordinates is as follows: , In the above formula, These are the vertex coordinates of the clipping path. The center coordinates of the packaging box blank These are the length and width of the packaging box, respectively.

[0059] S602. Apply heat-pressing film to the cut product; The surface is coated. The cut semi-finished product is conveyed to the coating unit. The coating temperature is set to 90℃, the hot pressing pressure is set to 0.3MPa, and the coating speed is consistent with the conveying speed. The thickness of the protective film on the surface of the coated product is 0.05mm.

[0060] S603. Remove surface impurities using an ultrasonic cleaning device; Activate the ultrasonic cleaning device, set the frequency to 40kHz, to remove dust, residual adhesive residue, and other impurities from the product surface.

[0061] S604. Count and pack the products; After finishing, the products are sent to the finished product collection unit, where the output is counted by a counting sensor. The products are sorted and packed into boxes in groups of 50, and an entry signal is sent after the boxes are packed.

[0062] In this invention, by pre-treating the raw materials, the deformation of the substrate caused by uneven moisture and poor flatness of the raw materials is effectively eliminated, providing a stable and qualified blank for subsequent high-precision hot stamping. By using fusion positioning, the three-dimensional spatial coordinates of the hot stamping target area are accurately identified, overcoming the shortcomings of traditional methods that cannot compensate for individual deformations, and improving the positional accuracy and consistency of the hot stamping pattern. By adaptively adjusting the hot stamping parameters, the hot stamping temperature, pressure, and speed are dynamically optimized based on a machine learning model, ensuring that the process parameters always match the characteristics of the raw materials and environmental conditions, avoiding hot stamping problems caused by improper parameters, and ensuring the optimization of the hot stamping effect and batch-to-batch stability.

[0063] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Clearly, those skilled in the art can make various alterations and variations to the invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of equivalents of the invention, the invention is also intended to include these modifications and variations.

Claims

1. A method for producing hot stamping packaging boxes, characterized in that, Includes the following steps: S1. Adjust the humidity and flatness of the raw material of the packaging box blank so that the humidity and flatness of the raw material reach the preset threshold. S2. Identify the target area for hot stamping on the blank and obtain its three-dimensional coordinates; S3. Based on raw material characteristics, environmental parameters, and positioning data, calculate hot stamping temperature, pressure, and speed parameters using a machine learning model; S4. Based on the three-dimensional coordinates and hot stamping parameters, control the hot stamping plate to adhere to the blank for hot stamping. S5. Perform defect detection on the hot-stamped blanks and sort qualified and unqualified products according to the detection results; S6. Cut, coat, remove impurities and collect qualified products.

2. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S1 includes the following steps: S101. Transfer the billet raw material to the processing area; S102. Collect the moisture content and flatness error value of the raw materials; S103. Compare the humidity value with the preset threshold, and start the humidifier or dehumidifier to adjust the humidity; S104. Adjust the pressure of the pressure rollers according to the flatness error value to correct the flatness. S105. Continuously monitor the data. If the standard is met, proceed to S2; otherwise, reprocess.

3. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S2 includes the following steps: S201, preheating industrial camera and laser rangefinder sensor; S202. Simultaneously capture images of the billet and collect height data; S203. Perform grayscale conversion, filtering, and edge detection processing on the image; S204. Match feature points based on the SIFT algorithm and calculate 3D coordinates by combining height data.

4. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S3 includes the following steps: S301. Collect data on raw material moisture content, flatness error, ambient temperature, ambient humidity, and positioning height, as well as historical production data; S302. Normalize the data; S303. Use the random forest regression algorithm to train the parameter prediction model, input real-time data into the model, and predict the hot stamping parameters. S304. Compare the predicted parameters with the safety threshold and output the final parameters.

5. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S4 includes the following steps: S401: Receive positioning coordinates and hot stamping parameters, adjust the hot stamping plate to the initial position and start heating; S402: Temperature is controlled by a PID temperature control algorithm, and the position of the hot stamping plate is adjusted according to the coordinates; S403. Adjust the pressure to the set value and control the hot stamping plate to descend at the set speed for hot stamping; S404. After hot stamping is completed, reset the hot stamping plate and transfer the blank.

6. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S5 includes the following steps: S501, Preheat the sensor and set the defect judgment threshold; S502. Collect the spectral image, RGB image, and gloss value of the hot stamping area; S503, Extract color, spectral and shape features; S504. Use a CNN model to classify defects and combine it with gloss to determine acceptance. S505. Based on the judgment results, product diversion shall be carried out.

7. The method for producing a hot stamping packaging box according to claim 1, characterized in that, S6 includes the following steps: S601. Laser cut the blank according to the preset dimensions; S602. Apply heat-pressing film to the cut product; S603. Remove surface impurities using an ultrasonic cleaning device; S604. Count and pack the products.

8. The method for producing a hot stamping packaging box according to claim 5, characterized in that, In S403, the hot stamping time is calculated based on pressure, temperature, and speed. The expression is as follows: In the above formula, For hot stamping time, Based on the hot stamping time, For the synergy coefficient, For correction factor, This refers to the hot stamping temperature. For hot stamping pressure, This refers to the speed of hot stamping.

9. A method for producing a hot stamping packaging box according to claim 7, characterized in that, In S601, the expression for calculating the cut path coordinates is as follows: In the above formula, These are the vertex coordinates of the clipping path. The center coordinates of the packaging box blank These are the length and width of the packaging box, respectively.

10. A hot stamping packaging box production system, applicable to the hot stamping packaging box production method according to any one of claims 1-9, characterized in that, include: The raw material pretreatment module is used to adjust the humidity and correct the flatness of the packaging box blank raw material; The fusion positioning module is used to identify the hot stamping target area of ​​the packaging box blank, obtain its coordinate information in three-dimensional space, and connect to the raw material pretreatment module; The hot stamping parameter adjustment module calculates the appropriate hot stamping temperature, pressure, and speed parameters based on raw material characteristics, environmental parameters, positioning data, and historical production data, and connects to the fusion positioning module. The hot stamping execution module is used to complete the bonding, hot stamping, and separation of the hot stamping plate and the packaging box blank according to the positioning data and parameters. It connects the hot stamping parameter adjustment module and the fusion positioning module. The quality inspection module is used to inspect the blanks after hot stamping, identify defect types, and sort the products. It is connected to the hot stamping execution module. The finished product post-processing module cuts, coats, and removes impurities from qualified hot stamping blanks to obtain finished hot stamping packaging boxes, which are then connected to the quality inspection module.

Images