Sample two-dimensional code intelligent printing system

Abstract

The invention provides a sample two-dimensional code intelligent printing system, and relates to the technical field of industrial printing, and the system comprises the steps: obtaining pressure data and surface image data of a test block sample, carrying out the fusion judgment, and obtaining a test block sample molding signal; constructing a test block sample digital twinborn model to obtain an actual compressive strength value; obtaining a predicted compressive strength value based on the neural convolutional network model; obtaining the minimum recognizable size of the two-dimensional code according to the three-dimensional point cloud data of the test block sample; adjusting the minimum recognizable size of the two-dimensional code to obtain a standard two-dimensional code recognition size; generating a two-dimensional code automatic planning code spraying path; on the basis of a two-dimensional code generation algorithm, the transition two-dimensional code is obtained and encrypted, a code spraying path is automatically planned according to the two-dimensional code, dynamic adjustment is carried out according to a standard two-dimensional code recognition size, then the corresponding test block sample two-dimensional code is generated, and efficiency and safety are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of industrial printing technology, and in particular to a sample two-dimensional code intelligent printing system. Background Art

[0002] Concrete and cement specimens are the core carriers of engineering quality inspection. Their unique identification and full-process traceability are key links to ensure structural safety. Traditional specimen identification technology mainly relies on physical label implantation. By fixing waterproof PVC or metal QR code labels in the mold, the specimens with identification are formed after pouring and curing. However, there are significant limitations:

[0003] 1. High dependence on manual labor: Label implantation requires manual positioning and fixation (such as using U-shaped metal cards or hot melt adhesive), which can easily cause labels to shift or fall off due to operational errors; test piece size verification and QR code size adjustment require manual measurement and parameter setting, which is inefficient and prone to errors.

[0004] 2. Insufficient environmental adaptability: Most inkjet printers are in fixed parameter mode and cannot automatically adjust the inkjet size according to the size of the test piece. Problems often occur where the QR code is too large to cover the test surface of the test piece or too small to be recognized.

[0005] 3. Limited utilization of data value: QR codes only store basic information (such as specimen number and molding time), and lack real-time connection with the maintenance environment and test data. Quality traceability relies on manual review of paper records, and cannot achieve intelligent early warning of detection anomalies and defect location.

[0006] Therefore, a sample QR code intelligent printing system is now provided. Summary of the Invention

[0007] In order to solve the above technical problems, the purpose of the present invention is to provide a sample QR code intelligent printing system.

[0008] In order to achieve the above object, the present invention provides the following technical solutions: A sample QR code intelligent printing system, comprising: An intelligent mold formation detection module is used to obtain pressure data and surface image data of the test block sample during molding; fuse and judge the pressure data and surface image data to obtain a mold formation signal of the test block sample; The test block sample model construction module obtains the three-dimensional point cloud data of the corresponding test block sample based on the test block sample modeling signal; based on the digital twin technology and the three-dimensional point cloud data, constructs the test block sample digital twin model to obtain the actual compressive strength value of the corresponding test block sample; based on the neural convolution network model and the actual compressive strength value of the test block sample, obtains the predicted compressive strength value of the corresponding test block sample; The QR code size planning module obtains the minimum recognizable size of the QR code based on the three-dimensional point cloud data of the corresponding test block sample; adjusts the minimum recognizable size of the QR code to obtain the standard QR code recognition size; The dynamic QR code generation module generates a QR code and automatically plans the inkjet printing path based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the maintenance environment data; based on the QR code generation algorithm, it obtains the transition QR code and encrypts the transition QR code, automatically plans the inkjet printing path based on the QR code, and dynamically adjusts the standard QR code recognition size to generate the corresponding test block sample QR code.

[0009] According to one preferred embodiment of the present invention, the process of obtaining pressure data and surface image data of a test block sample during molding includes: A data acquisition device is provided, including a pressure data acquisition unit, an image data acquisition unit, and a data acquisition unit, and an acquisition cycle is set, wherein the acquisition cycle includes a plurality of acquisition moments; a plurality of pressure data acquisition units are evenly distributed at the bottom of the molding die of the test block sample, wherein the pressure data acquisition units sense pressure data generated during the molding process of the test block according to the acquisition moments; An image data acquisition unit is set up just above the forming mold to photograph the surface of the test block sample according to different acquisition angles, acquisition light sources and acquisition times to obtain surface image data.

[0010] According to one preferred embodiment of the present invention, the process of fusing and judging the pressure data and the surface image data includes: Preset the pressure threshold range according to the material properties, size specifications and molding process requirements of the test block sample; Monitor the pressure data at the time of collection, and when the pressure data remains stable within the pressure threshold range When 1 second is reached, the molding pressure of the test block sample reaches the expected requirement, which is recorded as the implementation pressure data; based on The algorithm detects and processes the edges of surface image data. The process includes: The surface image data is converted into a grayscale image, denoted as ; The grayscale image Perform Gaussian smoothing to obtain a smooth grayscale image ; based on Algorithm for smooth grayscale images Perform convolution processing to obtain the corresponding gradient amplitude ; The preset gradient amplitude low threshold is , the high threshold value of the gradient amplitude is ; like , then the pixel is a strong edge point; if , then the pixel is a weak edge point; if , then the pixel is suppressed and identified as a defect on the edge of the test block sample; Based on the gray level co-occurrence matrix texture analysis algorithm, the smooth gray image Performing texture analysis to obtain corresponding texture feature parameters; the texture feature parameters include: contrast, correlation, energy, and homogeneity; According to the contrast, correlation, energy and homogeneity, the texture feature similarity is obtained. ; The texture feature similarity for: ;in, is contrast; for energy; For relevance; for homogeneity; 、 、 as well as is the corresponding weight coefficient; Set standard texture feature similarity ; like If , there are defects on the surface of the test block sample; if When , there is no defect on the surface of the test block sample; According to the implementation of pressure data, gradient amplitude and texture feature similarity After comprehensive judgment, the test block samples that meet the corresponding requirements will be used to generate test block sample molding signals.

[0011] According to one preferred embodiment of the present invention, the process of constructing a digital twin model of a test block sample based on three-dimensional point cloud data includes: Importing 3D point cloud data into digital twin modeling software, geometrically modeling the 3D point cloud data, and converting the discrete point cloud data into a continuous 3D model surface using a triangulation algorithm; associating the corresponding applied pressure data and surface image data with the model, and adding physical properties to the model based on actual needs; and establishing a digital twin model of the test block sample by simulating the physical behavior of the test block sample under different environmental conditions. Set curing conditions, import the curing conditions into the digital twin model of the test block sample as a test environment, perform a compressive strength test on the digital twin model of the test block sample, and obtain the actual compressive strength value of the test block sample.

[0012] According to one preferred embodiment of the present invention, based on the neural convolutional network model and according to the actual compressive strength value of the test block sample, the process of obtaining the predicted compressive strength value of the corresponding test block sample includes: Collect data from several groups of test block samples as training data sets, and normalize the training data sets; Input the preprocessed training data set into the neural convolutional network model for training; set the weight update formula and bias update formula of the neural convolutional network model; The weight update formula is: , the bias update formula ,in, is the learning rate; is the loss function, and the specific formula is: ,in, is the actual compressive strength value, To predict the compressive strength value; After the training of the neural convolutional network model is completed, the data corresponding to the sample to be pre-tested is input into the trained neural convolutional network model. The neural convolutional network model performs calculations based on the parameters obtained through training and outputs the predicted compressive strength value of the corresponding test block sample.

[0013] According to one preferred embodiment of the present invention, the process of obtaining the minimum recognizable size of a QR code based on the three-dimensional point cloud data of the corresponding test block sample includes: According to the roughness and reflective properties of the test block surface, a relationship model between the minimum recognizable size of the QR code and the surface characteristics of the test block is established, wherein the relationship model includes a surface roughness relationship model and a surface reflective property relationship model; The surface roughness relationship model construction process includes: Construct a surface roughness relationship model based on the local curvature of each point in the 3D point cloud data; Determine the point Neighborhood ,in, is the neighborhood radius; Based on the principal component analysis algorithm, the neighborhood Analyze and obtain three principal components 、 、 And the corresponding eigenvalues 、 、 ,and ; According to the three principal components 、 、 And the corresponding eigenvalues 、 、 , get points The local curvature of , the average roughness of the entire test block sample surface ; According to the average roughness , constructing a surface roughness relationship model ;in, is a positive coefficient; The surface reflection characteristic relationship model construction process includes: Set up a point The reflection intensity is , the average reflection intensity of the entire test block surface ; According to the average reflection intensity , build a surface reflective properties relationship model ;in, is the ideal reflection intensity; is a positive coefficient; Extract the surface size information of the test block sample from the 3D point cloud data; according to the surface size information of the test block sample and the scanning distance, obtain the initial recognizable size of the QR code and record it as ; According to the initial recognizable size , surface roughness relationship model and surface reflective properties relationship model , calculate the minimum recognizable size of the QR code .

[0014] According to one preferred embodiment of the present invention, the process of adjusting the minimum recognizable size of a QR code to obtain a standard QR code recognition size includes: Preset QR code size standard range ; If the minimum recognizable size of the QR code Smaller than the lower limit of the standard size range When the QR code is the smallest recognizable size Adjust to the lower limit ; When the minimum recognizable size of the QR code Larger than the upper limit of the standard size range When the QR code is the smallest recognizable size Adjust to upper limit ; The minimum recognizable size of the adjusted QR code based on the information capacity To update, the information stored in the preset QR code is , calculate the minimum number of coding units required based on the information content ; The size space occupied by each coding unit is recorded as , to obtain the minimum size ; The minimum size The minimum recognizable size of the QR code after the current adjustment For comparison: If , you need to increase the size of the QR code; according to and The difference , set the current minimum recognizable size of the QR code to Updated to ; like , then the minimum recognizable size of the currently adjusted QR code is The information storage requirements are met and no further adjustments are required; The minimum recognizable size of the adjusted QR code , recorded as the standard QR code recognition size.

[0015] According to one preferred embodiment of the present invention, the process of generating a QR code and automatically planning a coding path based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the curing environment data includes: The digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample and the curing environment data are integrated to obtain the information tuple to be encoded , based on UTF-8 encoding, the information tuple to be encoded Convert to a string format of QR code encoding; According to the surface roughness relationship model , determine the starting area of ​​coding and make sure the coding direction is horizontal; Construct a grid map of the test block sample surface; Discretize the coding surface of the test block sample into a two-dimensional grid map , each grid cell Represents a small area on the coding surface; For each grid cell Assign a status, including: passable and impassable; Each grid cell As a node , the node Contains location information , parent node pointer, actual cost from the starting node to this node And the estimated cost from this node to the target node Total cost ; According to the actual cost , estimated cost and the total cost , generate a QR code and automatically plan the coding path. The process includes: Based on the A* algorithm, preset open list and closed list; The starting node Add to said open list and add its actual price Set to , calculate the estimated cost and the total cost , the closed list is initially empty; 1. Select a node: select the total cost from the open list Smallest node ; 2. Check target: If the node is the target node , then the search ends, and the coding path is obtained by backtracking through the parent node pointer, which is recorded as the QR code automatic planning coding path; 3. Mark explored: mark the node Remove from the open list and add to the closed list; 4. Traverse adjacent nodes: If the node Not the target node , traverse the nodes All adjacent nodes ; 5. Ignore invalid nodes: If the adjacent nodes If the node is in the closed list or is inaccessible, the node is ignored; 6. Calculate the new actual cost: Calculate from the node To adjacent nodes The new actual cost ; 7. Update node information: If the adjacent node Not in the open list or Smaller than adjacent nodes Current actual cost , update the adjacent nodes The actual cost , estimated cost , total cost and the parent node pointer, and the adjacent nodes Add to open list until adjacent nodes Target node , the search ends, and the coding path is obtained by backtracking through the parent node pointer and recorded as the QR code automatic planning coding path.

[0016] According to one preferred embodiment of the present invention, the process of automatically planning the coding path based on the QR code and dynamically adjusting the standard QR code recognition size to generate the corresponding test block sample QR code includes: The information tuple to be encoded Input into the QR code generation algorithm and encode the information tuple Encode into a binary matrix, where each matrix element corresponds to a module of the QR code, 0 represents a white module, and 1 represents a black module, generating a transitional QR code; Encrypting the generated transitional QR code to obtain an encrypted QR code and an encryption key, and decoding the encrypted QR code based on a QR code decoding tool and the encryption key to obtain a decoding result; Obtain an encrypted QR code whose decoding result meets the standard requirements, automatically plan the coding path and standard QR code recognition size based on the QR code, use the coding equipment to print the encrypted QR code on the surface of the test block sample, and obtain the corresponding test block sample QR code.

[0017] A computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement the sample QR code intelligent printing system.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. The system uses an intelligent mold detection module to acquire pressure data and surface image data, integrating and judging them to accurately determine whether the test block has been molded. Compared to traditional manual judgment or single-data judgment, it integrates multi-source data and considers multiple aspects of the test block's molding, improving the accuracy and reliability of mold determination. In complex molding environments, single pressure data may not accurately reflect the actual molding of the test block. However, combined with surface image data, it can identify surface defects and other problems that pressure data cannot reflect, thereby preventing unqualified test blocks from entering subsequent processes, improving the quality and efficiency of the entire production process, and reducing costs caused by test block quality issues.

[0019] 2. Based on digital twin technology and neural convolutional network models, actual and predicted compressive strength values ​​can be obtained. This enables the system to not only obtain actual compressive strength values ​​by simulating the physical behavior of test blocks under different environmental conditions using the digital twin model, but also predict compressive strength values ​​using the neural convolutional network model. In this way, the compressive strength of test blocks can be evaluated and predicted in advance, providing a basis for parameter adjustments during the production process. Based on the predicted compressive strength values, the curing conditions or material ratios of the test blocks can be adjusted to improve the quality of the test blocks and meet the compressive strength requirements of different projects, while also providing stronger support for project quality control.

[0020] 3. Based on the roughness, reflective properties, surface size information, and scanning distance of the test block surface, a relational model is established to calculate the minimum recognizable size of the QR code. The size is adjusted by presetting the standard QR code size range and considering the information capacity to obtain the standard QR code recognition size. The dynamic QR code generation module integrates the information to be encoded based on the test block related data, plans the coding path based on the A* algorithm, generates a transition QR code and encrypts it, and dynamically adjusts the path and size to generate the final test block sample QR code. This adaptive QR code generation and coding path planning technology takes into account the various characteristics and actual needs of the test block surface, so that the generated QR code can better adapt to the test block surface, ensure the recognizability and data security of the QR code, and optimize the coding path to improve coding efficiency and quality. It is an innovation and improvement in QR code generation and coding technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0022] Figure 1 A schematic diagram of the steps of a sample QR code intelligent printing system.

[0023] Figure 2 This is a module diagram of a sample QR code intelligent printing system. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.

[0025] like Figure 2 As shown, a sample QR code intelligent printing system includes: An intelligent mold formation detection module is used to obtain pressure data and surface image data of the test block sample during molding; fuse and judge the pressure data and surface image data to obtain a mold formation signal of the test block sample; The test block sample model construction module obtains the three-dimensional point cloud data of the corresponding test block sample based on the test block sample modeling signal; based on the digital twin technology and the three-dimensional point cloud data, constructs the test block sample digital twin model to obtain the actual compressive strength value of the corresponding test block sample; based on the neural convolution network model and the actual compressive strength value of the test block sample, obtains the predicted compressive strength value of the corresponding test block sample; The QR code size planning module obtains the minimum recognizable size of the QR code based on the three-dimensional point cloud data of the corresponding test block sample; adjusts the minimum recognizable size of the QR code to obtain the standard QR code recognition size; The dynamic QR code generation module generates a QR code and automatically plans the inkjet printing path based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the maintenance environment data; based on the QR code generation algorithm, it obtains the transition QR code and encrypts the transition QR code, automatically plans the inkjet printing path based on the QR code, and dynamically adjusts the standard QR code recognition size to generate the corresponding test block sample QR code.

[0026] It should be further explained that, in the specific implementation process, the specific process of obtaining the pressure data and surface image data of the test block sample during molding includes: A data acquisition device is provided, including a pressure data acquisition unit, an image data acquisition unit, and a data acquisition unit, and an acquisition cycle is set, wherein the acquisition cycle includes a plurality of acquisition moments; a plurality of pressure data acquisition units are evenly distributed at the bottom of the molding die of the test block sample, wherein the pressure data acquisition unit senses the pressure data generated during the molding process of the test block according to the acquisition moment, and transmits the pressure data to the data acquisition unit in the form of an electrical signal; the data acquisition unit performs analog-to-digital conversion and preliminary processing on the electrical signal, and then sends the pressure data to the background system.

[0027] An image data acquisition unit is set up directly above the forming mold to photograph the surface of the test block sample according to different acquisition angles, acquisition light sources and acquisition times to obtain surface image data; the surface image data is transmitted to the background system through the image transmission interface.

[0028] It should be further explained that, in order to ensure the clarity and accuracy of the surface image data, the image data acquisition unit is equipped with a suitable light source and lens, and can automatically adjust the exposure parameters under different ambient lighting conditions.

[0029] For example, the pressure data acquisition unit in the data acquisition device includes but is not limited to a high-precision pressure sensor, and the image data acquisition unit includes but is not limited to a high-definition camera, and the specific settings are based on actual needs.

[0030] It should be further explained that, in the specific implementation process, the specific process of fusing and judging the pressure data and the surface image data includes: Preset the pressure threshold range according to the material properties, size specifications and molding process requirements of the test block sample; For example, for a concrete specimen of a certain strength grade, the pressure threshold range during the forming process may be set as ,in, is the lower limit value, These thresholds are determined through a large number of experiments and actual production experience, and can ensure that the test block is formed under appropriate pressure conditions.

[0031] Monitor the pressure data at the time of collection, and when the pressure data remains stable within the pressure threshold range When 1 second is reached, the molding pressure of the test block sample reaches the expected requirement, which is recorded as the implementation pressure data; It should be further explained that by analyzing the time series of pressure data, we can judge whether the pressure data is rising, falling or stable. The fluctuation of pressure data within a time period of seconds does not exceed a certain range. If the pressure remains within the range for t seconds and complies with the pressure variation law of the specimen molding, such as the pressure gradually rises during the specimen filling stage and is relatively stable during the initial setting stage, the pressure data is considered normal and meets the molding pressure requirements, and ΔP is the allowable pressure fluctuation value.

[0032] based on The algorithm detects and processes the edges of surface image data. The specific process includes: The surface image data is converted into a grayscale image, denoted as ; The grayscale image Perform Gaussian smoothing to obtain a smooth grayscale image , the smooth grayscale image for: ;in, is a two-dimensional Gaussian function, and Is a grayscale image The coordinates of the pixel point; based on Algorithm for smooth grayscale images Perform convolution processing to obtain the corresponding gradient amplitude , the gradient amplitude for: ; The preset gradient amplitude low threshold is , the high threshold value of the gradient amplitude is ; like , then the pixel is a strong edge point; if , then the pixel is a weak edge point; if , then the pixel is suppressed and identified as a defect on the edge of the test block sample; It should be further explained that the edge contour of the test block sample surface is identified, and whether the shape of the test block is complete is determined by detecting the position and shape of the test block sample edge. The strong edge point is definitely an edge point, and a weak edge point is only considered an edge point when it is connected to a strong edge point.

[0033] For example, for a cubic concrete specimen, its edges should be clear and regular, and edge detection can be used to detect whether there are any missing corners or damage.

[0034] Based on the gray level co-occurrence matrix texture analysis algorithm, the smooth gray image Performing texture analysis to obtain corresponding texture feature parameters; the texture feature parameters include: contrast, correlation, energy, and homogeneity; Suppose smooth grayscale image The gray level is , the distance is , the angle is , gray-level co-occurrence matrix , Indicates the gray value is The pixel point is at a distance of , angle is The gray value is The frequency of pixels; The contrast ;in, is the total number of pixels excluding edge pixels; The correlation ;in, ; ; ; ; The energy ; The homogeneity ; According to the contrast ,energy , correlation and homogeneity , obtain texture feature similarity ; The texture feature similarity for: ;in, 、 、 as well as is the corresponding weight coefficient; Set standard texture feature similarity ; like If , there are defects on the surface of the test block sample; if When , there are no defects on the surface of the test block sample.

[0035] It should be further explained that, based on the local binary texture analysis algorithm, the surface flatness and uniformity of the test block sample are detected, and the texture characteristic parameters of the test block sample surface, such as texture roughness and directionality, can be calculated. If the test block sample surface has obvious texture unevenness or defects (such as honeycombing or pockmarks), the texture characteristic parameters will change accordingly, thus determining whether the test block surface is abnormal.

[0036] According to the implementation of pressure data, gradient amplitude and texture feature similarity After comprehensive judgment, the test block samples that meet the corresponding requirements will be used to generate test block sample molding signals.

[0037] It should be further explained that, in the specific implementation process, the specific process of constructing the digital twin model of the test block sample based on the 3D point cloud data includes: When the mold signal of the test block sample is received, the three-dimensional laser scanner is started to scan the test block sample, and the three-dimensional point cloud data of each point on the test block surface is calculated by measuring the reflection time and angle of the laser beam.

[0038] It should be further explained that during the scanning process, the 3D laser scanner scans the test block from multiple angles to obtain complete 3D information. The raw data obtained from the scan is then preprocessed, such as removing noise points and filtering data, to obtain accurate 3D point cloud data.

[0039] The acquired three-dimensional point cloud data is imported into the digital twin modeling software, and the three-dimensional point cloud data is geometrically modeled. The discrete point cloud data is converted into a continuous three-dimensional model surface using a triangulation algorithm. The corresponding applied pressure data and surface image data are associated with the model, and physical properties such as material properties and density are added to the model according to actual needs. By simulating the physical behavior of the test block sample under different environmental conditions, a digital twin model of the test block sample is established.

[0040] It should be further explained that pressure data is used to simulate the stress conditions of the test block during the molding process, and surface image data is used to verify that the surface features of the model are consistent with the actual test block. By continuously adjusting the model parameters, the digital twin model of the test block sample can accurately reflect the actual state of the test block sample.

[0041] Curing conditions are set and imported into the digital twin model of the test block sample as a test environment. The compressive strength test of the digital twin model of the test block sample is performed until the digital twin model of the test block sample is destroyed, and the actual compressive strength value of the test block sample is calculated.

[0042] It should be further explained that, in the specific implementation process, based on the neural convolutional network model and according to the actual compressive strength value of the test block sample, the specific process of obtaining the predicted compressive strength value of the corresponding test block sample includes: Collect data from several groups of test block samples, including sample size, material ratio, curing environment, actual compressive strength, etc., as training data sets. Normalize the training data sets to convert the data to the same scale range.

[0043] Based on the neural convolutional network model, the model includes input layer, convolution layer, pooling layer, fully connected layer and output layer; It should be further explained that the preprocessed training data set is input into the model for training, and the error between the model output and the actual compressive strength value is minimized by continuously adjusting the model weights and biases; Set the weight update formula and bias update formula of the neural convolutional network model to update the neural convolutional network model and reduce the error; The weight update formula is: , the bias update formula ,in, is the learning rate; is the loss function, and the specific formula is: ,in, is the actual compressive strength value, To predict the compressive strength value.

[0044] After the training of the neural convolutional network model is completed, the data corresponding to the sample to be pre-tested is input into the trained neural convolutional network model. The neural convolutional network model performs calculations based on the parameters obtained through training and outputs the predicted compressive strength value of the corresponding test block sample.

[0045] It should be further explained that, in the specific implementation process, the specific process of obtaining the minimum recognizable size of the QR code based on the three-dimensional point cloud data of the corresponding test block sample includes: Based on the roughness and reflective properties of the test block surface, a relationship model between the minimum recognizable size of the QR code and the surface characteristics of the test block is established. The relationship model includes a surface roughness relationship model and a surface reflective property relationship model. It should be further explained that the rough surface of the test block will affect the recognition effect of the QR code, and the size of the QR code needs to be appropriately increased.

[0046] The surface roughness relationship model construction process includes: Construct a surface roughness relationship model based on the local curvature of each point in the 3D point cloud data; Determine the point Neighborhood ,in, is the neighborhood radius; Based on the principal component analysis algorithm, the neighborhood Analyze and obtain three principal components 、 、 And the corresponding eigenvalues 、 、 ,and ; According to the three principal components 、 、 And the corresponding eigenvalues 、 、 , get points The local curvature of , so the average roughness of the entire specimen surface is ; According to the average roughness , constructing a surface roughness relationship model ;in, It is a positive coefficient, determined by the experimental process; The surface reflection characteristic relationship model construction process includes: Set up a point The reflection intensity is , the average reflection intensity of the entire test block surface ; According to the average reflection intensity , build a surface reflective properties relationship model ;in, is the ideal reflection intensity; It is a positive coefficient, determined by the experimental process; Extract the surface dimension information of the test block sample from the 3D point cloud data, including length, width, height, etc. According to the surface dimension information of the test block sample and the scanning distance, obtain the initial recognizable size of the QR code and record it as ; It should be further explained that the farther the scanning distance, the larger the size of the QR code needs to be to ensure accurate recognition.

[0047] According to the initial recognizable size , surface roughness relationship model and surface reflective properties relationship model , calculate the minimum recognizable size of the QR code .

[0048] It should be further explained that, in the specific implementation process, the minimum recognizable size of the QR code is adjusted to obtain the standard QR code recognition size. The specific process includes: Preset QR code size standard range ; When the calculated minimum recognizable size of the QR code Smaller than the lower limit of the standard size range In order to ensure the recognizability and compliance of the QR code, the minimum recognizable size of the QR code is Adjust to the lower limit , that is: if When ; When the minimum recognizable size of the QR code Larger than the upper limit of the standard size range When the QR code is the smallest recognizable size Adjust to upper limit , that is: If When ; The minimum recognizable size of the adjusted QR code based on the information capacity To update, assume that the information stored in the QR code is , calculate the minimum number of coding units required based on the information content ; Each coding unit occupies a certain size space, recorded as , then theoretically the minimum size of information storage is satisfied ; Will The minimum recognizable size of the QR code after the current adjustment For comparison: If , you need to increase the size of the QR code; according to and The difference , set the current minimum recognizable size of the QR code to Updated to ; like , then the minimum recognizable size of the currently adjusted QR code is The information storage requirements are met and no further adjustments are required; The minimum recognizable size of the adjusted QR code , recorded as the standard QR code recognition size.

[0049] It should be further explained that, during the specific implementation process, the specific process of generating a QR code and automatically planning the coding path based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the curing environment data includes: The digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample and the curing environment data are integrated to obtain the information tuple to be encoded , based on UTF-8 encoding, the information tuple to be encoded Convert to a string format of QR code encoding; According to the surface roughness relationship model , determine the starting area of ​​coding and make sure the coding direction is horizontal; It should be further explained that the roughness of the surface of the test block sample is detected to find relatively flat areas on the surface. These areas are more suitable as the starting areas for coding.

[0050] Construct a grid map of the test block sample surface; Discretize the coding surface of the test block sample into a two-dimensional grid map , each grid cell Represents a small area on the coding surface; it should be further explained that the grid unit The size needs to be determined according to the size of the QR code and the coding accuracy to ensure that each grid unit can accommodate one coding unit of the QR code.

[0051] For each grid cell Assign a status, including: passable (indicating that the area is suitable for coding) and impassable (indicating that the area has obstacles or uneven surfaces and is not suitable for coding), and mark the impassable grid cells .

[0052] Each grid cell As a node , the node Contains location information , parent node pointer (for backtracking path), actual cost from the starting node to the node And the estimated cost from this node to the target node ; It should be further explained that the actual cost Indicates that from the starting node To the current node For example, in a grid map, the actual cost between adjacent nodes can be set to 1 (horizontally or vertically) or (diagonal movement); the estimated cost , based on the Manhattan distance algorithm to estimate from the current node To the target node For example, the Manhattan distance ,in, is the coordinate of the target node; the total cost ; According to the actual cost , estimated cost and the total cost , generate a QR code and automatically plan the coding path. The specific process includes: Based on the A* algorithm, preset open list and closed list; The starting node Add to said open list and add its actual price Set to , calculate the estimated cost and the total cost , the closed list is initially empty; 13. Select Node: Select the total cost from the open list Smallest node ; 14. Check target: If the node is the target node , then the search ends, and the coding path is obtained by backtracking through the parent node pointer, which is recorded as the QR code automatic planning coding path; 15. Mark explored: mark the node Removed from the open list and added to the closed list.

[0053] 16. Traverse adjacent nodes: If the node Not the target node , traverse the nodes All adjacent nodes ; 17. Ignore invalid nodes: If the adjacent nodes If a node is in the closed list or its status is inaccessible, the node is ignored.

[0054] 18. Calculate the new actual cost: Calculate from the node To adjacent nodes The new actual cost .

[0055] 19. Update node information: If the adjacent node Not in the open list or Smaller than adjacent nodes Current actual cost , update the adjacent nodes The actual cost , estimated cost , total cost and the parent node pointer, and the adjacent nodes Add to open list until adjacent nodes Target node , the search ends, and the coding path is obtained by backtracking through the parent node pointer and recorded as the QR code automatic planning coding path.

[0056] It should be further explained that, in the specific implementation process, the specific process of automatically planning the coding path based on the QR code and dynamically adjusting the standard QR code recognition size to generate the corresponding test block sample QR code includes: The information tuple to be encoded Input into the QR code generation algorithm and encode the information tuple Encode into a binary matrix, where each matrix element corresponds to a module of the QR code, 0 represents a white module, and 1 represents a black module, generating a transitional QR code; Encrypting the generated transitional QR code to obtain an encrypted QR code and an encryption key; decoding the encrypted QR code using a QR code decoding tool and the encryption key to obtain a decoding result; and verifying whether the integrated information can be accurately read. The decoding result meets the standard requirements and can accurately reflect the encrypted QR code of the test block sample related information, and automatically plans the coding path and the standard QR code recognition size according to the QR code, and uses the coding equipment to print the encrypted QR code on the surface of the test block sample to obtain the corresponding test block sample QR code.

[0057] It should be further explained that, in the specific implementation process, the specific process of encrypting the generated transitional QR code includes: Encrypt the generated transition QR code based on the symmetric encryption algorithm; Randomly generate an encryption key and securely store the encryption key in a background system; encrypt the encoding data of the transitional two-dimensional code according to the encryption key, and store the encrypted encoding data of the transitional two-dimensional code in a storage module of the inkjet printer; During the coding process, the coding device transmits the encrypted two-dimensional code to the print head, and the print head prints the encrypted two-dimensional code on the surface of the test block; During the code scanning and identification process, the same encryption key is used to decrypt the QR code to obtain the information contained therein, ensuring that the data of the test block sample cannot be tampered with.

[0058] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A sample QR code intelligent printing system, characterized in that: include: Intelligent mold detection module, used to obtain pressure data and surface image data of test block samples during molding; Performing fusion judgment on the pressure data and the surface image data, thereby obtaining a mold forming signal of the test block sample; The test block sample model building module obtains the three-dimensional point cloud data of the corresponding test block sample according to the test block sample modeling signal; Based on the digital twin technology and according to the three-dimensional point cloud data, a digital twin model of the test block sample is constructed to obtain the actual compressive strength value of the corresponding test block sample; based on the neural convolution network model and according to the actual compressive strength value of the test block sample, a predicted compressive strength value of the corresponding test block sample is obtained; The QR code size planning module obtains the minimum recognizable size of the QR code based on the three-dimensional point cloud data of the corresponding test block sample; adjusts the minimum recognizable size of the QR code to obtain the standard QR code recognition size; The dynamic QR code generation module generates a QR code and automatically plans the inkjet printing path based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the maintenance environment data; based on the QR code generation algorithm, it obtains the transition QR code and encrypts the transition QR code, automatically plans the inkjet printing path based on the QR code, and dynamically adjusts the standard QR code recognition size to generate the corresponding test block sample QR code.

2. A sample QR code intelligent printing system according to claim 1, characterized in that: The process of obtaining pressure data and surface image data during the molding of the test block sample includes: A data acquisition device is provided, including a pressure data acquisition unit, an image data acquisition unit, and a data acquisition unit, and an acquisition cycle is set, wherein the acquisition cycle includes a plurality of acquisition moments; a plurality of pressure data acquisition units are evenly distributed at the bottom of the molding die of the test block sample, wherein the pressure data acquisition units sense pressure data generated during the molding process of the test block according to the acquisition moments; An image data acquisition unit is set up just above the forming mold to photograph the surface of the test block sample according to different acquisition angles, acquisition light sources and acquisition times to obtain surface image data.

3. A sample two-dimensional code intelligent printing system according to claim 2, characterized in that: The process of fusing and judging pressure data and surface image data includes: Preset the pressure threshold range according to the material properties, size specifications and molding process requirements of the test block sample; Monitor the pressure data at the time of collection, and when the pressure data remains stable within the pressure threshold range When 1 second is reached, the molding pressure of the test block sample reaches the expected requirement, which is recorded as the implementation pressure data; based on The algorithm detects and processes the edges of surface image data. The process includes: The surface image data is converted into a grayscale image, denoted as ; The grayscale image Perform Gaussian smoothing to obtain a smooth grayscale image ; based on Algorithm for smooth grayscale images Perform convolution processing to obtain the corresponding gradient amplitude ; The preset gradient amplitude low threshold is , the high threshold value of the gradient amplitude is ; like , then the pixel is a strong edge point; if , then the pixel is a weak edge point; if , then the pixel is suppressed and identified as a defect on the edge of the test block sample; Based on the gray level co-occurrence matrix texture analysis algorithm, the smooth gray image Performing texture analysis to obtain corresponding texture feature parameters; the texture feature parameters include: contrast, correlation, energy, and homogeneity; According to the contrast, correlation, energy and homogeneity, the texture feature similarity is obtained. ; The texture feature similarity for: ;in, is contrast; for energy; For relevance; for homogeneity; 、 、 as well as is the corresponding weight coefficient; Set standard texture feature similarity ; like If , there are defects on the surface of the test block sample; if When , there is no defect on the surface of the test block sample; According to the implementation of pressure data, gradient amplitude and texture feature similarity After comprehensive judgment, the test block samples that meet the corresponding requirements will be used to generate test block sample molding signals.

4. A sample two-dimensional code intelligent printing system according to claim 3, characterized in that: The process of building a digital twin model of a test block sample based on 3D point cloud data includes: Importing 3D point cloud data into digital twin modeling software, geometrically modeling the 3D point cloud data, and converting the discrete point cloud data into a continuous 3D model surface using a triangulation algorithm; associating the corresponding applied pressure data and surface image data with the model, and adding physical properties to the model based on actual needs; and establishing a digital twin model of the test block sample by simulating the physical behavior of the test block sample under different environmental conditions. Set curing conditions, import the curing conditions into the digital twin model of the test block sample as a test environment, perform a compressive strength test on the digital twin model of the test block sample, and obtain the actual compressive strength value of the test block sample.

5. A sample two-dimensional code intelligent printing system according to claim 4, characterized in that: The process of obtaining the predicted compressive strength value of the corresponding test block sample based on the neural convolutional network model and the actual compressive strength value of the test block sample includes: Collect data from several groups of test block samples as training data sets, and normalize the training data sets; Input the preprocessed training data set into the neural convolutional network model for training; set the weight update formula and bias update formula of the neural convolutional network model; The weight update formula is: , the bias update formula ,in, is the learning rate; is the loss function, and the specific formula is: ,in, is the actual compressive strength value, To predict the compressive strength value; After the training of the neural convolutional network model is completed, the data corresponding to the sample block to be pre-tested is input into the trained neural convolutional network model. The neural convolutional network model performs calculations based on the parameters obtained through training and outputs the predicted compressive strength value of the corresponding test block sample.

6. A sample two-dimensional code intelligent printing system according to claim 5, characterized in that: The process of obtaining the minimum recognizable size of the QR code based on the 3D point cloud data of the corresponding test block sample includes: According to the roughness and reflective properties of the test block surface, a relationship model between the minimum recognizable size of the QR code and the surface characteristics of the test block is established, wherein the relationship model includes a surface roughness relationship model and a surface reflective property relationship model; The surface roughness relationship model construction process includes: Construct a surface roughness relationship model based on the local curvature of each point in the 3D point cloud data; Determine the point Neighborhood ,in, is the neighborhood radius; Based on the principal component analysis algorithm, the neighborhood Analyze and obtain three principal components 、 、 And the corresponding eigenvalues 、 、 ,and ; According to the three principal components 、 、 And the corresponding eigenvalues 、 、 , get points The local curvature of , the average roughness of the entire test block sample surface ; According to the average roughness , constructing a surface roughness relationship model ;in, is a positive coefficient; The surface reflection characteristic relationship model construction process includes: Set up a point The reflection intensity is , the average reflection intensity of the entire test block surface ; According to the average reflection intensity , build a surface reflective properties relationship model ;in, is the ideal reflection intensity; is a positive coefficient; Extract the surface size information of the test block sample from the 3D point cloud data; according to the surface size information of the test block sample and the scanning distance, obtain the initial recognizable size of the QR code and record it as ; According to the initial recognizable size , surface roughness relationship model and surface reflectivity relationship model , calculate the minimum recognizable size of the QR code .

7. A sample two-dimensional code intelligent printing system according to claim 6, characterized in that: The process of adjusting the minimum recognizable size of a QR code to obtain the standard QR code recognition size includes: Preset QR code size standard range ; If the minimum recognizable size of the QR code Smaller than the lower limit of the standard size range When the QR code is the smallest recognizable size Adjust to the lower limit ; When the minimum recognizable size of the QR code Larger than the upper limit of the standard size range When the QR code is the smallest recognizable size Adjust to upper limit ; The minimum recognizable size of the adjusted QR code based on the information capacity To update, the information stored in the preset QR code is , calculate the minimum number of coding units required based on the information content ; The size space occupied by each coding unit is recorded as , to obtain the minimum size ; The minimum size The minimum recognizable size of the QR code after the current adjustment For comparison: If , you need to increase the size of the QR code; according to and The difference , set the current minimum recognizable size of the QR code to Updated to ; like , then the minimum recognizable size of the currently adjusted QR code is The information storage requirements are met and no further adjustments are required; The minimum recognizable size of the adjusted QR code , recorded as the standard QR code recognition size.

8. A sample two-dimensional code intelligent printing system according to claim 7, characterized in that: The process of automatically planning the inkjet printing path by generating a QR code based on the digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample, and the curing environment data includes: The digital twin model of the test block sample, the actual compressive strength value, the predicted compressive strength value, the basic information of the test block sample and the curing environment data are integrated to obtain the information tuple to be encoded , based on UTF-8 encoding, the information tuple to be encoded Convert to a string format of QR code encoding; According to the surface roughness relationship model , determine the starting area of ​​coding and make sure the coding direction is horizontal; Construct a grid map of the test block sample surface; Discretize the coding surface of the test block sample into a two-dimensional grid map , each grid cell Represents a small area on the coding surface; For each grid cell Assign a status, including: passable and impassable; Each grid cell As a node , the node Contains location information , parent node pointer, actual cost from the starting node to this node And the estimated cost from this node to the target node Total cost ; According to the actual cost , estimated cost and the total cost , generate a QR code and automatically plan the coding path. The process includes: Based on the A* algorithm, preset open list and closed list; The starting node Add to said open list and add its actual price Set to , calculate the estimated cost and the total cost , the closed list is initially empty; Select Node: Select the total cost from the open list Smallest node ; Check target: If the node is the target node , then the search ends, and the coding path is obtained by backtracking through the parent node pointer, which is recorded as the QR code automatic planning coding path; Mark explored: Remove from the open list and add to the closed list; Traverse adjacent nodes: If the node Not the target node , traverse the nodes All adjacent nodes ; Ignore invalid nodes: If the adjacent nodes If the node is in the closed list or is inaccessible, the node is ignored; Calculate the new actual cost: Calculate from the node To adjacent nodes The new actual cost ; Update node information: If the adjacent node Not in the open list or Smaller than adjacent nodes Current actual cost , update the adjacent nodes The actual cost , estimated cost , total cost and the parent node pointer, and the adjacent nodes Add to open list until adjacent nodes Target node , the search ends, and the coding path is obtained by backtracking through the parent node pointer and recorded as the QR code automatic planning coding path.

9. A sample two-dimensional code intelligent printing system according to claim 8, characterized in that: The process of automatically planning the coding path based on the QR code and dynamically adjusting the standard QR code recognition size to generate the corresponding test block sample QR code includes: The information tuple to be encoded Input into the QR code generation algorithm and encode the information tuple Encode into a binary matrix, where each matrix element corresponds to a module of the QR code, 0 represents a white module, and 1 represents a black module, generating a transitional QR code; Encrypting the generated transitional QR code to obtain an encrypted QR code and an encryption key, and decoding the encrypted QR code based on a QR code decoding tool and the encryption key to obtain a decoding result; Obtain an encrypted QR code whose decoding result meets the standard requirements, automatically plan the coding path and standard QR code recognition size based on the QR code, use the coding equipment to print the encrypted QR code on the surface of the test block sample, and obtain the corresponding test block sample QR code.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement a sample two-dimensional code intelligent printing system as described in any one of claims 1 to 9.

Images