Corrugated board production quality regulation and control method and system based on machine vision

Abstract

The invention relates to a corrugated board production quality regulation and control method and system based on machine vision, and relates to the technical field of machine vision, and the method comprises the steps: obtaining an image of a corrugated board; based on the image of the corrugated board, generating an image detection result of the corrugated board; based on the image detection result of the corrugated board, performing laser detection on the corrugated board to obtain laser detection data of the corrugated board; based on the laser detection data of the corrugated board, generating a laser detection result of the corrugated board; based on the laser detection result of the corrugated board, performing ultrasonic detection on the corrugated board to obtain ultrasonic detection data of the corrugated board; based on the ultrasonic detection data of the corrugated board, generating an ultrasonic detection result of the corrugated board; according to the image detection result, the laser detection result and the ultrasonic detection result of the corrugated board, the fault equipment of the corrugated board production line and the fault reason of the fault equipment are determined, and the corrugated board production quality automatic regulation and control device has the advantage that the production quality of the corrugated board is automatically regulated and controlled.

Description

Technical Field

[0001] The invention relates to the technical field of machine vision, and in particular to a method and a system for controlling the production quality of corrugated paperboard based on machine vision. Background Art

[0002] At present, corrugated cardboard is recognized as a green and environmentally friendly packaging product worldwide. At the same time, with the increase in logistics volume, the demand for corrugated boxes is also increasing. Existing corrugated cardboard, also known as corrugated cardboard, is made of at least one layer of corrugated paper and one layer of boxboard paper. It has good elasticity and extensibility and is mainly used to make cartons, sandwich panels for cartons, and other packaging materials for fragile goods.

[0003] However, in the traditional corrugated cardboard production control scheme, it is usually dependent on manual inspection of corrugated cardboard, which requires a lot of human resources and time. In addition, manual inspection is easily affected by the subjective factors of the operator. Different operators may have different judgment criteria and preferences, resulting in inconsistent inspection results. The same batch of products may get different judgment results between different operators. With the expansion of production scale, the traditional manual inspection method faces the challenge of processing large-scale data and cannot meet the needs of high production.

[0004] Therefore, it is necessary to provide a corrugated cardboard production quality control method and control system based on machine vision, which can be used to realize automatic control of the corrugated cardboard production quality. Summary of the invention

[0005] The present invention provides a method for controlling the production quality of corrugated cardboard based on machine vision, comprising: acquiring an image of the corrugated cardboard; generating an image detection result of the corrugated cardboard based on the image of the corrugated cardboard; performing laser detection on the corrugated cardboard based on the image detection result of the corrugated cardboard to obtain laser detection data of the corrugated cardboard; generating a laser detection result of the corrugated cardboard based on the laser detection data of the corrugated cardboard; performing ultrasonic detection on the corrugated cardboard based on the laser detection result of the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard; generating an ultrasonic detection result of the corrugated cardboard based on the ultrasonic detection data of the corrugated cardboard; and determining faulty equipment of a corrugated cardboard production line and the fault cause of the faulty equipment according to the image detection result, the laser detection result and the ultrasonic detection result of the corrugated cardboard.

[0006] Furthermore, the image of the corrugated cardboard is obtained, including: obtaining the image of the corrugated cardboard through a structured light camera; generating an image detection result of the corrugated cardboard based on the image of the corrugated cardboard, including: extracting a corrugated cardboard area image from the image of the corrugated cardboard; for any two pixels of the corrugated cardboard area image, calculating the grayscale difference between the two pixels; dividing the pixels of the corrugated cardboard area image into a plurality of pixel units according to the image distance between any two pixels and the grayscale difference between any two pixels; for each pixel unit, calculating the grayscale difference parameter of the pixel unit according to the grayscale value of each pixel included in the pixel unit, and determining whether the pixel unit is a defective pixel unit according to the grayscale difference parameter of the pixel unit, wherein the image detection result of the corrugated cardboard includes the grayscale difference parameter of each defective pixel unit.

[0007] Furthermore, the pixels of the corrugated cardboard area image are divided into a plurality of pixel units according to the image distance between any two pixels and the grayscale difference between any two pixels, including: using a clustering algorithm, the pixels of the corrugated cardboard area image are divided into a plurality of pixel clusters according to the image distance between any two pixels; for each pixel cluster, the pixels included in the pixel cluster are divided into at least one pixel unit according to the grayscale difference between any two pixels included in the pixel cluster.

[0008] Furthermore, based on the image detection result of the corrugated cardboard, laser detection is performed on the corrugated cardboard to obtain laser detection data of the corrugated cardboard, including: determining the laser detection area of ​​the corrugated cardboard according to the defective pixel unit; determining the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard according to the grayscale difference parameter corresponding to the defective pixel unit; and obtaining the laser detection data of the corrugated cardboard according to the laser detection area of ​​the corrugated cardboard and the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard, wherein the laser detection data of the corrugated cardboard includes laser ranging values ​​of multiple laser detection positions in each laser detection area.

[0009] Furthermore, based on the laser detection data of the corrugated cardboard, a laser detection result of the corrugated cardboard is generated, including: for each laser detection area, according to the laser ranging values ​​of multiple laser detection positions in the laser detection area, a laser ranging difference parameter of the laser detection area is calculated; according to the laser ranging difference parameter of the laser detection area, whether the laser detection area is a defective laser detection area is determined, wherein the laser detection result of the corrugated cardboard includes a defective laser detection area and the laser ranging difference parameter of the defective laser detection area.

[0010] Furthermore, based on the image detection results and laser detection results of the corrugated cardboard, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard, including: determining the ultrasonic detection area of ​​the corrugated cardboard according to the defective laser detection area; determining the ultrasonic detection density of the ultrasonic detection area according to the laser ranging difference parameters of the defective laser detection area; and performing ultrasonic detection on the corrugated cardboard according to the ultrasonic detection density to obtain ultrasonic detection data of the corrugated cardboard, wherein the ultrasonic detection data includes ultrasonic detection waveforms of multiple ultrasonic detection positions in the ultrasonic detection area.

[0011] Furthermore, the ultrasonic detection density of the ultrasonic detection area is determined according to the laser ranging difference parameters of the defective laser detection area, including: obtaining test data of sample corrugated cardboard corresponding to the corrugated cardboard; determining the center laser ranging difference parameters according to the test data of the sample corrugated cardboard corresponding to the corrugated cardboard; determining the ultrasonic detection density of the ultrasonic detection area according to the laser ranging difference parameters of the defective laser detection area and the center laser ranging difference parameters.

[0012] Furthermore, based on the ultrasonic detection data of the corrugated cardboard, an ultrasonic detection result of the corrugated cardboard is generated, including: obtaining a standard ultrasonic waveform of the corrugated cardboard; for each ultrasonic detection position, calculating a waveform difference value between the ultrasonic detection waveform of the ultrasonic detection position and the standard ultrasonic waveform; averaging the waveform difference values ​​between the ultrasonic detection waveform of each ultrasonic detection position and the standard ultrasonic waveform to generate a waveform difference mean, wherein the ultrasonic detection result of the corrugated cardboard includes the waveform difference mean.

[0013] Furthermore, based on the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard, the faulty equipment of the corrugated cardboard production line and the cause of the faulty equipment are determined, including: establishing a defect and equipment failure association map; through a fault prediction model, based on the defect and equipment failure association map and the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard, the faulty equipment of the corrugated cardboard production line and the cause of the faulty equipment are determined.

[0014] The present invention provides a corrugated cardboard production quality control system based on machine vision, which applies the above-mentioned corrugated cardboard production quality control method based on machine vision, including: an image acquisition module, used to acquire an image of the corrugated cardboard; a production analysis module, used to generate an image detection result of the corrugated cardboard based on the image of the corrugated cardboard; a laser detection module, used to perform laser detection on the corrugated cardboard based on the image detection result of the corrugated cardboard, and obtain laser detection data of the corrugated cardboard; the production analysis module is also used to generate a laser detection result of the corrugated cardboard based on the laser detection data of the corrugated cardboard; an ultrasonic detection module, used to perform ultrasonic detection on the corrugated cardboard based on the laser detection result of the corrugated cardboard, and obtain ultrasonic detection data of the corrugated cardboard; the production analysis module is also used to generate an ultrasonic detection result of the corrugated cardboard based on the ultrasonic detection data of the corrugated cardboard; the production analysis module is also used to determine the faulty equipment of the corrugated cardboard production line and the fault cause of the faulty equipment according to the image detection result, laser detection result and ultrasonic detection result of the corrugated cardboard.

[0015] Compared with the prior art, the present specification provides a method and system for controlling the quality of corrugated cardboard production based on machine vision, which has at least the following beneficial effects:

[0016] 1. Through laser technology and machine vision, the defect characteristics of corrugated cardboard can be accurately measured. By using the penetration of ultrasound, defects inside the corrugated cardboard, such as delamination and voids, can be detected to further improve product quality. Comprehensive analysis of image detection results, laser detection results, and ultrasonic detection results can accurately determine which equipment on the corrugated cardboard production line has failed and the specific cause of the failure. Through real-time monitoring and data analysis, potential faults can be discovered in a timely manner, and measures can be taken in advance for early warning and troubleshooting to prevent the fault from further expanding and affecting production progress and product quality;

[0017] 2. By calculating the grayscale difference between any two pixels in the corrugated cardboard image and dividing the pixels into multiple pixel units, fine-grained detection of surface defects of corrugated cardboard can be achieved. This method can capture subtle surface changes, such as tiny scratches, dents or unevenness, thereby improving the accuracy of detection. Defective pixel units are determined based on the grayscale difference parameters of the pixel units, so that the detection process can adapt to corrugated cardboards with different lighting conditions, different materials and different colors, enhancing the robustness and versatility of the detection. Determining the laser detection area based on the defective pixel units in the image detection results can achieve targeted detection of potential defective areas. This not only reduces unnecessary detection time, but also improves detection efficiency. Adjusting the laser detection density according to the grayscale difference parameters corresponding to the defective pixel units can ensure more intensive detection in areas with more serious defects, thereby obtaining more key information in a limited time;

[0018] 3. Ultrasonic testing can penetrate the surface of corrugated cardboard and detect internal structural defects such as delamination, voids or internal cracks. This makes up for the shortcomings of image detection and laser detection in internal quality detection. Determining the ultrasonic detection area according to the defect laser detection area avoids unnecessary comprehensive detection, thereby saving detection time and resources. By adjusting the ultrasonic detection density according to the laser ranging difference parameters of the defect laser detection area, different detection can be ensured in different areas, thereby improving the sensitivity and accuracy of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0020] Figure 1 It is a flow chart of a method for controlling the production quality of corrugated paperboard based on machine vision shown in an embodiment of the present application;

[0021] Figure 2 It is a module diagram of a corrugated cardboard production quality control system based on machine vision shown in one embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.

[0023] Figure 1 FIG. 1 is a flow chart of a method for controlling the quality of corrugated cardboard production based on machine vision shown in an embodiment of the present application. Figure 1 As shown, a method for controlling the production quality of corrugated cardboard based on machine vision may include the following steps.

[0024] Step 110, obtaining an image of the corrugated cardboard.

[0025] In some embodiments, step 110 specifically includes:

[0026] Use a structured light camera to obtain images of corrugated cardboard.

[0027] Specifically, a structured light camera is a three-dimensional imaging device based on structured light technology. The basic principle is to project light with certain structural characteristics (such as laser stripes, Gray code, sinusoidal stripes, etc.) onto the object being photographed (such as corrugated cardboard) through a projector (such as a near-infrared laser). These lights will form different reflection patterns at different depths of the object being photographed. Subsequently, one or more cameras capture these reflection patterns and convert the changes in this structure into depth information.

[0028] Step 120: Generate an image detection result of the corrugated cardboard based on the image of the corrugated cardboard.

[0029] In some embodiments, step 120 specifically includes:

[0030] Extracting a corrugated cardboard area image from the corrugated cardboard image;

[0031] For any two pixels in the corrugated cardboard area image, calculate the grayscale difference between the two pixels;

[0032] According to the image distance between any two pixels and the grayscale difference between any two pixels, the pixels of the corrugated cardboard area image are divided into a plurality of pixel units;

[0033] For each pixel unit, a grayscale difference parameter of the pixel unit is calculated according to the grayscale value of each pixel included in the pixel unit, and whether the pixel unit is a defective pixel unit is determined according to the grayscale difference parameter of the pixel unit, wherein the image detection result of the corrugated cardboard includes the grayscale difference parameter of each defective pixel unit. For example, a pixel unit whose grayscale difference parameter is greater than a grayscale difference parameter threshold value can be determined as a defective pixel unit.

[0034] Specifically, the collected corrugated cardboard image is grayed out, and the color image is converted into a grayscale image to simplify the subsequent processing steps. Local adaptive binarization is performed, and a suitable threshold is set according to the texture characteristics of the pixel points in the image of the corrugated cardboard, and the image is converted into a binary image, that is, the pixel points in the image have only two colors, black and white. Through automatic detection, the main area containing the corrugated cardboard is extracted, and irrelevant information such as the background is excluded to obtain the corrugated cardboard area image.

[0035] In some embodiments, pixels of the corrugated cardboard area image are divided into a plurality of pixel units according to an image distance between any two pixels and a grayscale difference between any two pixels, including:

[0036] By using a clustering algorithm (e.g., a K-means clustering algorithm, a hierarchical clustering algorithm, etc.), pixels of the corrugated cardboard area image are divided into a plurality of pixel clusters according to an image distance between any two pixels;

[0037] For each pixel cluster, the pixels included in the pixel cluster are divided into at least one pixel unit according to the grayscale difference between any two pixels included in the pixel cluster. For example, through a clustering algorithm (for example, K-means clustering algorithm, hierarchical clustering algorithm, etc.), the pixels included in the pixel cluster are divided into at least one pixel unit according to the grayscale difference between any two pixels included in the pixel cluster.

[0038] The grayscale difference parameter of the pixel unit can be calculated according to the following formula:

[0039]

[0040] Among them, θ (i,gray) is the grayscale difference parameter of the i-th pixel unit, G n is the grayscale value of the n-th pixel included in the ith pixel unit, and N is the grayscale value of the n-th pixel included in the ith pixel unit.

[0041] Step 130: Based on the image detection result of the corrugated cardboard, laser detection is performed on the corrugated cardboard to obtain laser detection data of the corrugated cardboard.

[0042] In some embodiments, step 130 specifically includes:

[0043] According to the defective pixel unit, the laser detection area of ​​the corrugated cardboard is determined. For example, the image coordinates of the pixels included in the defective pixel unit are converted into world coordinates to determine the laser detection area of ​​the corrugated cardboard. Before the coordinate conversion, the structured light camera needs to be calibrated. The calibration process includes determining the intrinsic parameters (such as focal length, optical center position, etc.) and extrinsic parameters (such as the orientation of the structured light camera relative to the world coordinate system) of the structured light camera. This is usually done by taking an image of a set of points with known spatial coordinates (such as checkerboard corners) and using the image coordinates and spatial coordinates of these points to calculate the parameters of the structured light camera. Once the structured light camera is calibrated, the image coordinates can be converted into world coordinates using the parameters of the structured light camera. This process involves a series of mathematical transformations, including rotation, translation, and projection. The specific conversion formula may vary depending on the structured light camera model and calibration method. But the basic idea is to map each pixel on the image to a point in the three-dimensional world coordinate system through the intrinsic and extrinsic parameters of the structured light camera;

[0044] Determine the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard according to the grayscale difference parameter corresponding to the defective pixel unit, wherein the laser detection density can reflect the number of laser detection positions in the laser detection area of ​​the corrugated cardboard;

[0045] According to the laser detection area of ​​the corrugated cardboard and the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard, the laser detection data of the corrugated cardboard is obtained, wherein the laser detection data of the corrugated cardboard includes laser ranging values ​​of multiple laser detection positions in each laser detection area.

[0046] Specifically, the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard can be determined according to the following formula:

[0047]

[0048] Among them, ρ (i,laser)is the laser detection density corresponding to the i-th laser detection area of ​​the corrugated cardboard, θ (0,gray) is the preset grayscale difference parameter, ρ (0,laser) is the preset laser detection density, and [ ] is the rounding operation.

[0049] Step 140: Generate a laser detection result of the corrugated cardboard based on the laser detection data of the corrugated cardboard.

[0050] In some embodiments, step 140 specifically includes:

[0051] For each laser detection area, a laser ranging difference parameter of the laser detection area is calculated according to the laser ranging values ​​of multiple laser detection positions in the laser detection area, and whether the laser detection area is a defective laser detection area is determined according to the laser ranging difference parameter of the laser detection area, wherein the laser detection result of the corrugated cardboard includes the defective laser detection area and the laser ranging difference parameter of the defective laser detection area. For example, a laser detection area whose laser ranging difference parameter is greater than the laser ranging difference parameter threshold is determined as a defective laser detection area.

[0052] Specifically, the laser ranging difference parameter can be calculated according to the following formula:

[0053]

[0054] Among them, θ (i,distance) is the laser ranging difference parameter of the i-th laser detection area, G n is the grayscale value of the n-th pixel included in the ith pixel unit, and N is the grayscale value of the n-th pixel included in the ith pixel unit.

[0055] Step 150: Based on the laser detection result of the corrugated cardboard, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard.

[0056] In some embodiments, step 150 specifically includes:

[0057] Determine the ultrasonic detection area of ​​the corrugated cardboard according to the defect laser detection area, for example, use the non-defect laser detection area as the ultrasonic detection area;

[0058] Determine the ultrasonic detection density of the ultrasonic detection area according to the laser ranging difference parameter of the defect laser detection area;

[0059] According to the ultrasonic detection density, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard, wherein the ultrasonic detection data includes ultrasonic detection waveforms of multiple ultrasonic detection positions in the ultrasonic detection area.

[0060] In some embodiments, determining the ultrasonic detection density of the ultrasonic detection area according to the laser ranging difference parameter of the defect laser detection area includes:

[0061] Obtain test data of sample corrugated cardboard corresponding to the corrugated cardboard;

[0062] Determine the center laser ranging difference parameter according to the test data of the sample corrugated cardboard corresponding to the corrugated cardboard;

[0063] The ultrasonic detection density of the ultrasonic detection area is determined according to the laser ranging difference parameter of the defect laser detection area and the center laser ranging difference parameter.

[0064] Specifically, the sample corrugated cardboard corresponding to the corrugated cardboard may be a test corrugated cardboard having the same structure and composition as the corrugated cardboard. The states of different sample corrugated cardboards may be different, for example, the number of defects, the location of defects, the causes of defects, and the characteristics of defects (for example, size) may be different. Among them, the defects of the sample corrugated cardboard are caused by abnormal internal structures.

[0065] For each sample corrugated cardboard, each ultrasonic detection area of ​​the sample corrugated cardboard can be determined according to steps 110-140, and the laser ranging difference parameter of each ultrasonic detection area of ​​the sample corrugated cardboard can be calculated, and the laser ranging difference parameter of each ultrasonic detection area of ​​the sample corrugated cardboard can be averaged to obtain the average value of the laser ranging difference parameter of the sample corrugated cardboard.

[0066] The mean of the laser ranging difference parameters of each sample corrugated cardboard is averaged to obtain the central laser ranging difference parameter.

[0067] The ultrasonic detection density of the ultrasonic detection area can be determined according to the following formula:

[0068]

[0069] Among them, ρ ultrasonic is the ultrasonic detection density of the ultrasonic detection area, I is the total number of defect laser detection areas, θ (center,ultrasonic) is the center laser ranging difference parameter, ρ (0,ultrasonic) It is the preset ultrasonic detection density.

[0070] Step 160: Generate ultrasonic detection results of the corrugated cardboard based on the ultrasonic detection data of the corrugated cardboard.

[0071] In some embodiments, step 160 specifically includes:

[0072] Obtaining a standard ultrasonic waveform of a corrugated cardboard, wherein the standard ultrasonic waveform may be an ultrasonic waveform of a sample corrugated cardboard of qualified quality that has the same structure and composition as the corrugated cardboard;

[0073] For each ultrasonic detection position, a waveform difference value between the ultrasonic detection waveform of the ultrasonic detection position and the standard ultrasonic waveform is calculated;

[0074] The waveform difference values ​​between the ultrasonic detection waveform of each ultrasonic detection position and the standard ultrasonic waveform are averaged to generate a waveform difference mean, wherein the ultrasonic detection result of the corrugated cardboard includes the waveform difference mean.

[0075] Specifically, the ultrasonic detection waveform can be subjected to variational modal decomposition to generate modal components of the ultrasonic detection waveform, and the waveform characteristics of the ultrasonic detection waveform can be extracted, such as the frequency characteristics (e.g., center frequency, frequency bandwidth, etc.), amplitude characteristics (e.g., peak amplitude, average amplitude, etc.), phase characteristics (e.g., phase difference, etc.), energy characteristics, etc. of each modal component. The ultrasonic detection waveform is decomposed into a series of modal components (also called intrinsic mode functions IMF) with different frequencies and amplitudes. These modal components can better reveal the details and features in the signal, and they help to extract key information about the internal defects or structural characteristics of the corrugated cardboard.

[0076] Similarly, the standard ultrasonic waveform can be subjected to variational modal decomposition to generate modal components of the standard ultrasonic waveform and extract waveform features of the standard ultrasonic waveform.

[0077] For example, the waveform difference value between the ultrasonic detection waveform of the ultrasonic detection position and the standard ultrasonic waveform can be calculated according to the following formula:

[0078]

[0079] Among them, γ is the waveform difference between the ultrasonic detection waveform at the ultrasonic detection position and the standard ultrasonic waveform, F (current,k) is the kth waveform feature of the ultrasonic detection waveform, F (standard,k) is the kth waveform feature of the standard ultrasonic waveform, and K is the total number of waveform features.

[0080] Step 170, determining the faulty equipment of the corrugated board production line and the faulty cause of the faulty equipment according to the image detection result, laser detection result and ultrasonic detection result of the corrugated board.

[0081] In some embodiments, step 170 specifically includes:

[0082] Establish a correlation map between defects and equipment failures;

[0083] Through the fault prediction model, according to the defect and equipment fault correlation map and the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard production line, the faulty equipment and the fault cause of the faulty equipment are determined, wherein the fault prediction model can be a convolutional network model.

[0084] Specifically, collect historical fault data of corrugated cardboard production lines, including faulty equipment, fault time, fault cause, etc. Collect equipment defect records, including defect type, defect location, defect degree and other information. Collect maintenance records to understand the maintenance process, maintenance measures and maintenance results. Clean and organize the collected data to remove duplicate, erroneous or invalid information. Standardize the data to ensure consistency in data format, units, etc. Analyze the correlation between equipment defects and failures to determine which defects may cause which failures. Based on historical data and expert experience, establish a mapping relationship between defects and failures. Use graph building tools (such as Neo4j, Jena, etc.) to represent entities such as equipment, defects, failures and the relationship between them in the form of a graph. Add attributes to entities and relationships in the graph for more detailed descriptions and queries.

[0085] Figure 2 is a module diagram of a corrugated board production quality control system based on machine vision shown in an embodiment of the present application, such as Figure 2 As shown, a corrugated cardboard production quality control system based on machine vision may include an image acquisition module, a production analysis module, a laser detection module and an ultrasonic detection module.

[0086] An image acquisition module, used for acquiring an image of the corrugated cardboard;

[0087] A production analysis module, used for generating an image detection result of the corrugated cardboard based on the image of the corrugated cardboard;

[0088] A laser detection module is used to perform laser detection on the corrugated cardboard based on the image detection result of the corrugated cardboard, and obtain laser detection data of the corrugated cardboard;

[0089] The production analysis module is also used to generate laser detection results of corrugated cardboard based on the laser detection data of the corrugated cardboard;

[0090] An ultrasonic detection module is used to perform ultrasonic detection on the corrugated cardboard based on the laser detection result of the corrugated cardboard, and obtain ultrasonic detection data of the corrugated cardboard;

[0091] The production analysis module is also used to generate ultrasonic testing results of corrugated cardboard based on ultrasonic testing data of corrugated cardboard;

[0092] The production analysis module is also used to determine the faulty equipment of the corrugated cardboard production line and the faulty cause of the faulty equipment according to the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard.

[0093] A corrugated cardboard production quality control system based on machine vision can be used to execute a corrugated cardboard production quality control method based on machine vision, which will not be described in detail here.

[0094] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A method for controlling the production quality of corrugated cardboard based on machine vision, characterized in that: include: Get an image of corrugated cardboard; Based on the image of the corrugated cardboard, generate the image detection result of the corrugated cardboard; Based on the image detection result of the corrugated cardboard, laser detection is performed on the corrugated cardboard to obtain laser detection data of the corrugated cardboard; Generate laser detection results of corrugated cardboard based on laser detection data of corrugated cardboard; Based on the laser detection result of the corrugated cardboard, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard; Generate ultrasonic testing results of corrugated cardboard based on ultrasonic testing data of corrugated cardboard; According to the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard, the faulty equipment of the corrugated cardboard production line and the cause of the faulty equipment are determined.

2. The method for controlling the production quality of corrugated cardboard based on machine vision according to claim 1, characterized in that: Get images of corrugated cardboard, including: Using a structured light camera, an image of the corrugated cardboard is acquired; Based on the image of the corrugated cardboard, the image detection results of the corrugated cardboard are generated, including: Extracting a corrugated cardboard area image from the corrugated cardboard image; For any two pixels in the corrugated cardboard area image, calculate the grayscale difference between the two pixels; Dividing the pixels of the corrugated cardboard area image into a plurality of pixel units according to the image distance between any two pixels and the grayscale difference between any two pixels; For each pixel unit, a grayscale difference parameter of the pixel unit is calculated according to the grayscale value of each pixel included in the pixel unit, and whether the pixel unit is a defective pixel unit is determined according to the grayscale difference parameter of the pixel unit, wherein the image detection result of the corrugated cardboard includes the grayscale difference parameter of each defective pixel unit.

3. The method for controlling the production quality of corrugated cardboard based on machine vision according to claim 2, characterized in that: According to the image distance between any two pixels and the grayscale difference between any two pixels, the pixels of the corrugated cardboard area image are divided into a plurality of pixel units, including: By using a clustering algorithm, the pixels of the corrugated cardboard area image are divided into a plurality of pixel clusters according to the image distance between any two pixels; For each pixel cluster, the pixels included in the pixel cluster are divided into at least one pixel unit according to the grayscale difference between any two pixels included in the pixel cluster.

4. The method for controlling the production quality of corrugated cardboard based on machine vision according to claim 2, characterized in that: Based on the image detection results of the corrugated cardboard, laser detection is performed on the corrugated cardboard to obtain laser detection data of the corrugated cardboard, including: Determine the laser detection area of ​​the corrugated cardboard according to the defective pixel unit; Determine the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard according to the grayscale difference parameter corresponding to the defective pixel unit; Laser detection data of the corrugated cardboard is obtained according to the laser detection area of ​​the corrugated cardboard and the laser detection density corresponding to the laser detection area of ​​the corrugated cardboard, wherein the laser detection data of the corrugated cardboard includes laser ranging values ​​of multiple laser detection positions in each laser detection area.

5. The method for controlling the production quality of corrugated board based on machine vision according to claim 3, characterized in that: Based on the laser detection data of corrugated cardboard, the laser detection results of corrugated cardboard are generated, including: For each laser detection area, a laser ranging difference parameter of the laser detection area is calculated according to the laser ranging values ​​of multiple laser detection positions in the laser detection area, and whether the laser detection area is a defective laser detection area is determined according to the laser ranging difference parameter of the laser detection area, wherein the laser detection result of the corrugated cardboard includes a defective laser detection area and the laser ranging difference parameter of the defective laser detection area.

6. The method for controlling the production quality of corrugated board based on machine vision according to claim 5, characterized in that: Based on the image detection results and laser detection results of the corrugated cardboard, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard, including: Determine the ultrasonic inspection area of ​​the corrugated cardboard based on the defect laser inspection area; Determine the ultrasonic detection density of the ultrasonic detection area according to the laser ranging difference parameter of the defect laser detection area; According to the ultrasonic detection density, ultrasonic detection is performed on the corrugated cardboard to obtain ultrasonic detection data of the corrugated cardboard, wherein the ultrasonic detection data includes ultrasonic detection waveforms of multiple ultrasonic detection positions in the ultrasonic detection area.

7. The method for controlling the production quality of corrugated board based on machine vision according to claim 6, characterized in that: According to the laser ranging difference parameters of the defect laser detection area, the ultrasonic detection density of the ultrasonic detection area is determined, including: Obtain test data of sample corrugated cardboard corresponding to the corrugated cardboard; Determine the center laser ranging difference parameter according to the test data of the sample corrugated cardboard corresponding to the corrugated cardboard; The ultrasonic detection density of the ultrasonic detection area is determined according to the laser ranging difference parameter of the defect laser detection area and the center laser ranging difference parameter.

8. The method for controlling the production quality of corrugated board based on machine vision according to claim 6, characterized in that: Based on the ultrasonic test data of corrugated cardboard, the ultrasonic test results of corrugated cardboard are generated, including: Obtain standard ultrasonic waveform of corrugated cardboard; For each ultrasonic detection position, a waveform difference value between the ultrasonic detection waveform of the ultrasonic detection position and the standard ultrasonic waveform is calculated; The waveform difference values ​​between the ultrasonic detection waveform of each ultrasonic detection position and the standard ultrasonic waveform are averaged to generate a waveform difference mean, wherein the ultrasonic detection result of the corrugated cardboard includes the waveform difference mean.

9. A method for controlling the production quality of corrugated board based on machine vision according to any one of claims 1 to 8, characterized in that: According to the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard, the faulty equipment of the corrugated cardboard production line and the faulty equipment's faulty causes are determined, including: Establish a correlation map between defects and equipment failures; Through the fault prediction model, according to the defect and equipment fault correlation map and the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard production line, the faulty equipment and the cause of the faulty equipment are determined.

10. A corrugated cardboard production quality control system based on machine vision, characterized in that: A method for controlling the quality of corrugated cardboard production based on machine vision as described in any one of claims 1 to 9, comprising: An image acquisition module, used for acquiring an image of the corrugated cardboard; A production analysis module, used for generating an image detection result of the corrugated cardboard based on the image of the corrugated cardboard; A laser detection module is used to perform laser detection on the corrugated cardboard based on the image detection result of the corrugated cardboard, and obtain laser detection data of the corrugated cardboard; The production analysis module is also used to generate laser detection results of the corrugated cardboard based on the laser detection data of the corrugated cardboard; An ultrasonic detection module is used to perform ultrasonic detection on the corrugated cardboard based on the laser detection result of the corrugated cardboard, and obtain ultrasonic detection data of the corrugated cardboard; The production analysis module is also used to generate ultrasonic testing results of the corrugated paperboard based on the ultrasonic testing data of the corrugated paperboard; The production analysis module is also used to determine the faulty equipment and the faulty cause of the faulty equipment in the corrugated cardboard production line according to the image detection results, laser detection results and ultrasonic detection results of the corrugated cardboard.

Images