Visual detection method and system for surface quality of shaving board

Through the visual inspection system, the particleboard image data is automatically captured and analyzed, and the burr distribution coefficient and leveling coefficient are constructed. Combined with the quality prediction model, the problems of time-consuming, labor-consuming and mis-checking of traditional detection methods are solved, and the efficient and automated detection of particleboard surface quality is achieved. , improves testing efficiency and product quality.

CN118941516AInactive Publication Date: 2024-11-12LANGFANG AOSONG WOOD IND CO LTD
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202410961492.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional particleboard detection methods rely on manual inspection, which is time-consuming and labor-intensive, and are prone to missed inspections and missed inspections, making it difficult to meet the needs of modern industries for high-efficiency and high-precision inspections, especially in large-scale production, which is difficult to achieve efficient and automated online inspections.

Method used

A particleboard surface quality visual inspection system was designed, including position locking module, visual inspection module, burr detection module, plate quality analysis module and feedback module. Image data is captured through multiple groups of visual inspection instruments, preprocessing and feature extraction are carried out, and the construction is carried out. Glut distribution coefficient and flatness coefficient are combined with the quality prediction model to obtain defect evaluation index to realize automated screening.

Benefits of technology

Improve the stability and consistency of detection, quickly identify and position burrs, reduce detection errors, improve detection efficiency and decision-making accuracy, realize multi-dimensional automated detection of particleboard surface quality, and reduce manual inspection workload and error rate , reduces the scrap rate and rework rate, and improves the product's pass rate and market competitiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941516A_ABST
    Figure CN118941516A_ABST
Patent Text Reader

Abstract

The invention discloses a shaving board surface quality visual detection method and system, and relates to the technical field of visual detection, a three-dimensional coordinate system is established, to-be-detected shaving boards are placed on a workbench in advance, it is ensured that the positions of the shaving boards are consistent in the detection process, and detection errors caused by position deviation are avoided; and the detection stability and consistency are improved. The method comprises the following steps: capturing image data of each board surface of the shaving board by utilizing a plurality of groups of visual inspection instruments, pre-processing, constructing a burr distribution coefficient Mfxs according to burr state data, and comparing and analyzing the burr distribution coefficient Mfxs, so that a burr concentration area can be quickly identified and positioned, early warning can be timely given out, the possibility that the burr problem affects the product quality is reduced, and the product quality is improved. After an early warning instruction is received, a flatness coefficient Pzxs is constructed according to related flatness data on each panel, and a defect evaluation index Qpzs is comprehensively analyzed through key parameters, so that a scientific basis is provided for quality analysis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of visual inspection, and in particular to a particle board surface quality visual inspection method and system. Background Art

[0002] With the continuous development of intelligent manufacturing and industry, visual inspection technology has been widely used in the quality control of various industrial products. In the wood processing industry, particleboard, as a common composite board, is widely used in furniture manufacturing and architectural decoration. In the manufacturing process of particleboard, ensuring that its surface is burr-free, flat and smooth can not only improve the appearance and service life of the product, but also reduce the complexity and cost of subsequent processing. Therefore, visual inspection of particleboard surface quality has become an important research direction and application field in the industry.

[0003] Traditional particleboard inspection methods mostly rely on manual inspection, which is not only time-consuming and labor-intensive, but also prone to missed detections and false detections, making it difficult to meet the modern industry's demand for high-efficiency and high-precision inspections.

[0004] Specifically in the field of particleboard surface visual inspection, existing inspection technologies still have many shortcomings in practical applications. Traditional particleboard inspection methods mostly rely on manual inspection, which is not only time-consuming and labor-intensive, but also prone to missed inspections and false inspections. In addition, when identifying defects such as burrs on the particleboard surface and bubbles generated by the bonding between the particles and glue in the particleboard, they often lack systematicity and comprehensiveness, resulting in low accuracy and reliability of the inspection results. In addition, existing inspection systems are difficult to achieve efficient and automated online inspection when facing large-scale production. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a particleboard surface quality visual inspection method and system, which solves the problems in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a particleboard surface quality visual inspection system, including a position locking module, a visual inspection module, a burr detection module, a board quality analysis module and a feedback module;

[0007] The position locking module is used to place the particle board to be tested on the workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particle board to be tested;

[0008] The visual inspection module is used to capture the image data of each board surface of the particleboard to be inspected by using multiple sets of visual inspection instruments, and preprocess the image data of several sets of board surfaces to remove noise, smooth the image, and perform geometric correction, and then perform feature extraction on the preprocessed image data to respectively obtain relevant flatness data and relevant burr state data on each board surface to construct a visual inspection data set;

[0009] The burr detection module is used to construct the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, and if the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued externally;

[0010] The board quality analysis module is used to extract the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb according to the relevant flatness data on each board surface after receiving the early warning instruction, and obtain the bonding factor Njyz by correlating the bubble density Qpmd with the glue water content Jhz, and obtain the flatness coefficient Pzxs by correlating the bonding factor Njyz with the hole area ratio Dcmb, and obtain the defect evaluation index Qpzs by fitting in combination with the trained quality prediction model;

[0011] The feedback module is used to pre-set the evaluation threshold Q, and compare and analyze it with the defect evaluation index Qpzs, so as to comprehensively analyze the current state of the particleboard surface and perform screening operations.

[0012] Preferably, the position locking module comprises a plate position locking unit;

[0013] The plate locking unit is used to install a fixing device on the workbench, place the particleboard to be tested on the workbench through the fixing device, and use a laser rangefinder to measure the height of the particleboard relative to the workbench, and obtain the position coordinates of the particleboard to be tested by establishing a three-dimensional coordinate system, wherein the origin of the three-dimensional coordinate system is set at the exact center position of the particleboard to be tested.

[0014] Preferably, the visual detection module includes an image acquisition unit, a preprocessing unit and a feature extraction unit;

[0015] The image acquisition unit is used to install several groups of sensors and several groups of visual inspection instruments around the workbench, and after the particle board position is locked, start the several groups of visual inspection instruments to capture the image data of each board surface of the particle board to be inspected;

[0016] The preprocessing unit is used to convert the image of each board surface into a grayscale image and use a median filter to remove noise in the image;

[0017] The feature extraction unit is used to extract key feature information in the image through image enhancement technology, identify defective parts in the image, and respectively obtain relevant flatness data and relevant burr state data on each board surface, wherein the relevant flatness data includes the bubble density Qpmd on each surface of the particle board, the glue water content Jhz in the particle board production process, the glue curing time Jgsc in the particle board production process, the hole area ratio Dcmb and the maximum hole depth Kdsd;

[0018] The relevant burr status data includes the burr height Mcgd on the particleboard surface, and combined with statistical algorithms, the standard deviation of the burr height Mcgd is calculated. std and the mean value of burr height Mcgd avg .

[0019] Preferably, the burr detection module includes a preliminary analysis unit and a preliminary warning unit;

[0020] The preliminary analysis unit is used to analyze and calculate the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, which is specifically obtained according to the following formula:

[0021]

[0022] The significance of this formula is to: evaluate the discrete degree of burr height on the particleboard surface;

[0023] Among them, Mcgd std Expressed as the standard deviation of burr height, Mcgd avg Expressed as the mean burr height.

[0024] Preferably, the preliminary warning unit is used to pre-set a distribution threshold value, and compare and analyze it with the burr distribution coefficient Mfxs of the particle board to preliminarily analyze the current burr state of the particle board surface. The specific analysis content is as follows:

[0025] If the burr distribution coefficient Mfxs of the particleboard exceeds the distribution threshold, it will be preliminarily judged that the burrs on the particleboard surface are in an abnormal state and an early warning instruction will be issued;

[0026] If the burr distribution coefficient Mfxs of the particleboard does not exceed the distribution threshold, it will be preliminarily determined that the burrs on the surface of the particleboard are not in an abnormal state.

[0027] Preferably, the plate quality analysis module includes an interlayer penetration analysis unit, a surface analysis unit and a comprehensive analysis unit;

[0028] The interlayer penetration analysis unit is used to obtain the bonding factor Njyz after linear normalization of the bubble density Qpmd and the glue water content Jhz in the relevant flattening data on each board surface, which is specifically obtained according to the following formula:

[0029]

[0030] Where n is the number of particleboard surfaces, i = 1, 2, 3, ..., n, Qpmd i is expressed as the bubble density of the ith plate surface, It is represented by the average density of bubbles on the particleboard surface, Jgsc is represented by the glue curing time, and a1, a2 and a3 are all weight values.

[0031] Preferably, the surface analysis unit is used to calculate and obtain the flattening coefficient Pzxs based on the relevant flattening data on each board surface and in combination with the bonding factor Njyz after linear normalization processing, which is specifically obtained according to the following formula:

[0032]

[0033] Where Dcmb represents the proportion of hole area, Kdsd represents the maximum hole depth, C represents the first correction constant, and α and β are weight values.

[0034] Preferably, the comprehensive analysis unit is used to transmit the flattening coefficient Pzxs and the burr distribution coefficient Mfxs to the quality prediction model based on the visual inspection data set, and obtain the defect evaluation index Qpzs after linear normalization processing, which is specifically obtained according to the following formula:

[0035]

[0036] Wherein, R represents the second correction constant, and F1 and F2 are weight values.

[0037] Preferably, the feedback module includes an evaluation unit and a screening unit;

[0038] The evaluation unit is used to compare and analyze the evaluation threshold Q with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface. The specific comparison content is as follows:

[0039] If the defect evaluation index Qpzs is greater than the evaluation threshold Q, it indicates that the particleboard currently being tested is in an unqualified state, and a first result is generated at this time;

[0040] If the defect evaluation index Qpzs is equal to the evaluation threshold Q, it means that the particleboard currently being tested is in an unqualified state, and a second result is generated at this time;

[0041] If the defect evaluation index Qpzs is less than the evaluation threshold Q, it indicates that the particleboard currently being tested is in a qualified state, and a third result is generated at this time;

[0042] The screening unit is used to screen the particleboards corresponding to the first result, the second result and the third result, automatically screening the particleboards corresponding to the first result to the unqualified sorting area, automatically screening the particleboards corresponding to the second result to the repair sorting area, and automatically screening the particleboards corresponding to the third result to the qualified sorting area;

[0043] Among them, the particleboards in the sorting area to be repaired will be manually inspected, and the surface of the particleboard will be polished with sandpaper or a grinder to remove burrs. The filler should match the color and texture of the particleboard. For areas with abnormal bonding layers, glue will be re-applied and pressed under pressure.

[0044] A particleboard surface quality visual inspection method comprises the following steps:

[0045] Step 1: Place the particleboard to be tested on a workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particleboard to be tested;

[0046] Step 2: Then use multiple sets of visual inspection instruments to capture the image data of each board surface of the particleboard to be inspected, and pre-process several sets of board surface image data to remove noise, smooth the image, and perform geometric correction. Then, perform feature extraction on the pre-processed image data to obtain relevant flatness data and relevant burr state data on each board surface, so as to construct a visual inspection data set;

[0047] Step 3: Then, based on the relevant burr status data, the burr distribution coefficient Mfxs of the particleboard is constructed. If the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued;

[0048] Step 4. After receiving the early warning instruction, the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb are extracted according to the relevant flatness data on each board surface. The bonding factor Njyz is obtained by correlating the bubble density Qpmd with the glue water content Jhz. The flatness coefficient Pzxs is obtained by correlating the bonding factor Njyz with the hole area ratio Dcmb. Combined with the trained quality prediction model, the defect assessment index Qpzs is obtained by fitting.

[0049] Step 5. Finally, the evaluation threshold Q is set, and compared with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface and perform screening operations.

[0050] The present invention provides a particleboard surface quality visual inspection method and system, which has the following beneficial effects:

[0051] (1) By establishing a three-dimensional coordinate system, the particleboard to be tested is placed on the workbench in advance to ensure that each particleboard is in the same position during the test process, avoiding detection errors caused by position offset, and improving the stability and consistency of the test. Multiple sets of visual inspection instruments are used to capture the image data of each surface of the particleboard and preprocess it. The burr distribution coefficient Mfxs is constructed based on the burr status data, and compared and analyzed. This helps to quickly identify and locate the burr concentration area, and issue an early warning in time to reduce the possibility of burr problems affecting product quality. After receiving the early warning instruction, the flatness coefficient Pzxs is constructed based on the relevant flatness data on each board surface. By comprehensively analyzing these key parameters, the defect evaluation index Qpzs is obtained, providing a scientific basis for quality analysis. The evaluation threshold Q is set in advance and compared and analyzed with the defect evaluation index Qpzs. The current state of the particleboard surface is comprehensively analyzed. According to the comparison results, the corresponding screening operation is automatically generated to classify the particleboard as qualified, to be repaired, or unqualified, which improves the detection efficiency and decision accuracy. Through the organic combination of the above modules, the system can further realize multi-dimensional, automated detection and screening of the surface quality of particleboard, which not only reduces the workload and error rate of manual inspection, improves the work efficiency of the production line, but also reduces the scrap rate and rework rate, further improving the product's qualification rate and market competitiveness.

[0052] (2) The burr detection module can systematically evaluate the burr status of the particleboard surface through the cooperation of the preliminary analysis unit and the preliminary warning unit. Combined with the flatness data, the system can provide more comprehensive quality control and evaluation. Timely identification and processing of burr abnormalities can prevent unqualified particleboard from flowing into subsequent production links and reduce scrap and rework rates.

[0053] (3) Comprehensive analysis unit: Based on the visual inspection data set, the defect assessment index Qpzs is calculated, and the defect assessment index Qpzs is compared and analyzed according to the set evaluation threshold Q to generate the qualified or unqualified status result for the current particleboard. The screening unit automatically sorts the particleboard into different areas according to the results: unqualified sorting area, repair sorting area and qualified sorting area. Processing of the repair sorting area: For the particleboard sorted to the repair area, the system takes refined repair measures for different types of surface defects through manual inspection. Through the real-time feedback and refined repair process of the system, the occurrence rate of unqualified products is effectively reduced, and the production efficiency and product quality stability are improved. Qualified particleboards can be sorted directly to the shipping area, reducing secondary processing and scrap rate, and optimizing the operating efficiency and resource utilization of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a block diagram of a particleboard surface quality visual inspection system of the present invention;

[0055] Figure 2 The present invention is a schematic flow chart of a particleboard surface quality visual inspection method. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Example 1

[0058] See also Figure 1 , the present invention provides a particleboard surface quality visual inspection system, including a position locking module, a visual inspection module, a burr inspection module, a board quality analysis module and a feedback module;

[0059] The position locking module is used to place the particle board to be tested on the workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particle board to be tested;

[0060] The visual inspection module is used to capture the image data of each board surface of the particleboard to be inspected by using multiple sets of visual inspection instruments, and preprocess the image data of several sets of board surfaces to remove noise, smooth the image, and perform geometric correction, and then perform feature extraction on the preprocessed image data to respectively obtain relevant flatness data and relevant burr state data on each board surface to construct a visual inspection data set;

[0061] The burr detection module is used to construct the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, and if the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued externally;

[0062] The board quality analysis module is used to extract the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb according to the relevant flatness data on each board surface after receiving the early warning instruction, and obtain the bonding factor Njyz by correlating the bubble density Qpmd with the glue water content Jhz, and obtain the flatness coefficient Pzxs by correlating the bonding factor Njyz with the hole area ratio Dcmb, and obtain the defect evaluation index Qpzs by fitting in combination with the trained quality prediction model;

[0063] The feedback module is used to pre-set the evaluation threshold Q, and compare and analyze it with the defect evaluation index Qpzs, so as to comprehensively analyze the current state of the particleboard surface and perform screening operations.

[0064] During the operation of this system, firstly, through the position locking module, the particleboard to be inspected can be further accurately placed on the workbench, and a three-dimensional coordinate system can be established to lock the position coordinates of the particleboard. This can ensure that the position of the particleboard remains stable during the inspection process, thereby improving the accuracy and consistency of the inspection. Secondly, the visual inspection module uses multiple sets of visual inspection instruments to capture the image data of each board surface of the particleboard and perform preprocessing, including noise removal, image smoothing and geometric correction. Through feature extraction, the relevant flatness data and relevant burr state data of each board surface are obtained respectively to construct a complete visual inspection data set. This method can further improve the multi-facetedness and accuracy of the inspection, and can more accurately identify and quantify various defects on the surface of the particleboard. The burr detection module constructs the burr distribution coefficient Mfxs based on the relevant burr state data and sets the distribution threshold. When the burr distribution coefficient Mfxs exceeds the threshold, an early warning instruction is automatically issued. This early warning mechanism enables the system to discover and report potential quality problems in real time, take corrective measures in time, and further reduce the number of unqualified products entering the subsequent links. After receiving the early warning instruction, the board quality analysis module calculates the bonding factor Njyz and the flattening coefficient Pzxs based on the flattening data on each board surface, and then combines the trained quality prediction model to fit the defect assessment index Qpzs. This comprehensive analysis method can evaluate the overall quality of the particleboard in many aspects and provide a scientific basis for subsequent quality control. Finally, the feedback module pre-sets the evaluation threshold Q, compares it with the defect assessment index Qpzs, comprehensively analyzes the current state of the particleboard surface, and automatically screens according to the results. The design of this module enables the system to automatically sort the particleboard into three categories: qualified, unqualified, and to be repaired, which improves production efficiency and further reduces the labor intensity and misjudgment rate of manual inspection.

[0065] Example 2

[0066] Please refer to Figure 1 , specifically: the position locking module includes a plate position locking unit;

[0067] The board locking unit is used to install a fixing device on the workbench, wherein the fixing device includes but is not limited to a clamp, a vacuum suction cup and a positioning pin to ensure that the particleboard does not move during the detection process; the fixing device should be designed to be adjustable to accommodate particleboards of different sizes; the particleboard to be detected is placed on the workbench through the fixing device, and a laser rangefinder is used to measure the height of the particleboard relative to the workbench. The laser rangefinder can provide high-precision position information to help establish a three-dimensional coordinate system; by establishing the three-dimensional coordinate system, the position coordinates of the particleboard to be detected are obtained, wherein the origin of the three-dimensional coordinate system is set at the exact center of the particleboard to be detected.

[0068] The visual detection module includes an image acquisition unit, a preprocessing unit and a feature extraction unit;

[0069] The image acquisition unit is used to install several groups of sensors and several groups of visual inspection instruments around the workbench, and after the particle board position is locked, start the several groups of visual inspection instruments to capture the image data of each board surface of the particle board to be inspected;

[0070] The pre-processing unit is used to convert the image of each board surface into a grayscale image to reduce the processing complexity, and use median filtering to remove noise in the image to improve the image quality;

[0071] The feature extraction unit is used to extract key feature information in the image through image enhancement technology, identify defective parts in the image, and respectively obtain relevant flatness data and relevant burr status data on each board surface, wherein the relevant flatness data includes the bubble density Qpmd on each surface of the particle board, the glue moisture content Jhz in the particle board production process, the glue curing time Jgsc in the particle board production process, the hole area ratio Dcmb and the maximum hole depth Kdsd; wherein the glue moisture content Jhz can be monitored and obtained by a resistive humidity sensor; the glue curing time Jgsc can be monitored and obtained by a time controller.

[0072] The relevant burr status data includes the burr height Mcgd on the particleboard surface, and combined with statistical algorithms, the standard deviation of the burr height Mcgd is calculated. std and the mean value of burr height Mcgd avg .

[0073] In this embodiment, the plate position locking unit: a fixture, a vacuum suction cup, a positioning pin and other fixing devices are installed on the workbench, and a laser rangefinder is used to provide high-precision position information to establish a three-dimensional coordinate system. This module ensures that the particleboard will not move during the detection process, reduces the detection error caused by position offset, and can be adjusted to adapt to particleboards of different sizes, thereby improving the versatility and stability of the system. Image acquisition unit: By installing several groups of sensors and visual inspection instruments around the workbench, the image data of each surface of the particleboard is fully captured to ensure the comprehensiveness and accuracy of the data. Preprocessing unit: The image is converted into a grayscale image, and the median filter is used to remove noise, thereby improving the image quality and processing efficiency, and providing high-quality basic data for subsequent feature extraction. Feature extraction unit: The key feature information is extracted through image enhancement technology, the defective parts in the image are identified, and the flatness data and burr status data of each board surface are obtained. This module can accurately identify multiple groups of key parameters and reflect the surface quality of the particleboard in multiple directions.

[0074] Example 3

[0075] Please refer to Figure 1 ,Specifically: the burr detection module includes a preliminary analysis unit and a preliminary warning unit;

[0076] The preliminary analysis unit is used to analyze and calculate the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, which is specifically obtained according to the following formula:

[0077]

[0078] The significance of this formula is to: evaluate the discrete degree of burr height on the particleboard surface;

[0079] Among them, Mcgd std Expressed as the standard deviation of burr height, Mcgd avg Expressed as the mean burr height.

[0080] In this embodiment, by calculating the burr distribution coefficient Mfxs, the discrete degree of burr height on the particleboard surface can be evaluated, further reflecting the uniformity and severity of burr distribution. This burr distribution analysis helps to timely discover and locate burr problems on the particleboard surface, further improving the accuracy and comprehensiveness of detection. According to the burr distribution coefficient Mfxs, it is compared with the preset distribution threshold, and an early warning instruction is automatically issued. This early warning mechanism can timely identify abnormal conditions of burrs on the particleboard surface to prevent burr problems from affecting the overall quality of the particleboard. The burr detection module can systematically evaluate the burr state on the particleboard surface through the cooperation of the preliminary analysis unit and the preliminary early warning unit, and provide key data support for the subsequent board quality analysis module. Through automated burr detection and early warning, the system can further improve the quality control efficiency in the particleboard production process, reduce the workload and errors of manual inspection; the timely early warning mechanism helps to quickly take corrective measures, reduce the scrap rate in production, and improve the overall quality and market competitiveness of particleboard products.

[0081] Example 4

[0082] Please refer to Figure 1 Specifically: the preliminary warning unit is used to pre-set the distribution threshold value, and compare and analyze it with the burr distribution coefficient Mfxs of the particle board to preliminarily analyze the current burr state of the particle board surface. The specific analysis content is as follows:

[0083] If the burr distribution coefficient Mfxs of the particleboard exceeds the distribution threshold, it will be preliminarily judged that the burrs on the particleboard surface are in an abnormal state and an early warning instruction will be issued;

[0084] If the burr distribution coefficient Mfxs of the particleboard does not exceed the distribution threshold, it will be preliminarily determined that the burrs on the surface of the particleboard are not in an abnormal state.

[0085] In this embodiment, the burr distribution coefficient Mfxs of the particleboard is compared and analyzed with the threshold value to preliminarily analyze the current burr state of the particleboard surface. This accurate burr state evaluation method can further quickly identify the abnormal situation of the burr on the particleboard surface through the comparative analysis of the quantitative index Mfxs and the preset threshold value, thereby improving the accuracy and reliability of the detection. Early warning mechanism: During the detection process, once the burr distribution coefficient Mfxs exceeds the preset distribution threshold, the system can immediately issue an early warning instruction to prompt the operator or the system to take corresponding corrective measures. This timely early warning mechanism helps to quickly respond to the abnormal situation of the burr on the particleboard surface, prevent the further spread of quality problems, and ensure the continuity of the production process and the stability of product quality. Automated detection and early warning: Through the combination of the preliminary analysis unit and the preliminary early warning unit, the automation and intelligence of burr detection are realized. The system can automatically analyze and judge the burr state without manual intervention, and issue early warning instructions in time, which greatly improves the detection efficiency and reduces the workload and subjective errors of manual inspection. Improve production efficiency: Timely warning instructions can help the production line quickly identify and correct quality problems, reduce the scrap rate in production, and improve the overall quality and market competitiveness of particleboard products. Through accurate burr status assessment and timely abnormal warning mechanism, the system significantly improves the accuracy and efficiency of particleboard surface quality detection, improves the quality control ability in the production process, and ensures the high quality and market competitiveness of particleboard products.

[0086] Example 5

[0087] Please refer to Figure 1 ,Specifically: the plate quality analysis module includes an interlayer penetration analysis unit, a surface analysis unit and a comprehensive analysis unit;

[0088] The interlayer penetration analysis unit is used to obtain the bonding factor Njyz after linear normalization of the bubble density Qpmd and the glue water content Jhz in the relevant flattening data on each board surface, which is specifically obtained according to the following formula:

[0089]

[0090] Where n is the number of particleboard surfaces, i = 1, 2, 3, ..., n, Qpmd i is expressed as the bubble density of the ith plate surface, It is represented as the mean bubble density on the particleboard surface, Jgsc is represented as the glue curing time, a1, a2 and a3 are all weight values, among which, 0<a1≤1, 0<a2≤1, 0<a3≤1, and a1+a2+a3=1.

[0091] The surface analysis unit is used to calculate the flatness coefficient Pzxs according to the relevant flatness data on each board surface and the bonding factor Njyz after linear normalization processing. The calculation is performed according to the following formula:

[0092]

[0093] Wherein, Dcmb represents the proportion of hole area, Kdsd represents the maximum hole depth, C represents the first correction constant, α and β are weight values, where 0<α≤1, 0<β≤1, and α+β=1.

[0094] In this embodiment, the bonding quality of the particleboard is comprehensively evaluated by considering parameters such as the bubble density and glue water content of multiple board surfaces. This process can effectively quantify the bonding consistency between the board surfaces and further improve the accuracy of the overall quality judgment of the particleboard. Combined with the flatness data and bonding factor Njyz of each board surface, the flatness coefficient Pzxs is calculated. This coefficient comprehensively considers the hole area ratio Dcmb and the maximum hole depth Kdsd, thereby evaluating the flatness of the particleboard surface from multiple aspects. Through this analysis unit, the quality status of the particleboard surface can be further accurately reflected, providing a basis for subsequent quality control and improvement. The system effectively improves the quality control capability and product quality in the particleboard production process, and provides strong technical support and guarantee for enterprises to achieve efficient production and high-quality products.

[0095] Example 6

[0096] Please refer to Figure 1 Specifically: the comprehensive analysis unit is used to transmit the flattening coefficient Pzxs and the burr distribution coefficient Mfxs to the quality prediction model based on the visual inspection data set, and obtain the defect evaluation index Qpzs after linear normalization processing, which is specifically obtained according to the following formula:

[0097]

[0098] In the formula, R represents the second correction constant, F1 and F2 are weight values, wherein 0<F1≤1, 0<F2≤1, and F1+F2=1.

[0099] The feedback module includes an evaluation unit and a screening unit;

[0100] The evaluation unit is used to compare and analyze the evaluation threshold Q with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface. The specific comparison content is as follows:

[0101] If the defect evaluation index Qpzs is greater than the evaluation threshold Q, it indicates that the particleboard currently being tested is in an unqualified state, and a first result is generated at this time;

[0102] If the defect evaluation index Qpzs is equal to the evaluation threshold Q, it means that the particleboard currently being tested is in an unqualified state, and a second result is generated at this time;

[0103] If the defect evaluation index Qpzs is less than the evaluation threshold Q, it indicates that the particleboard currently being tested is in a qualified state, and a third result is generated at this time;

[0104] The screening unit is used to screen the particleboards corresponding to the first result, the second result and the third result, automatically screening the particleboards corresponding to the first result to the unqualified sorting area, automatically screening the particleboards corresponding to the second result to the repair sorting area, and automatically screening the particleboards corresponding to the third result to the qualified sorting area;

[0105] Among them, the particleboards in the sorting area to be repaired will be subject to manual inspection, and the surface of the particleboard will be polished with sandpaper or a grinder to remove burrs. Sandpaper of different coarseness and fineness can be selected according to the severity of the burrs, and the surface will be gradually polished until it is smooth; for areas with depressions on the surface, wood fillers or resins will be used to fill them. The fillers should match the color and texture of the particleboard to ensure the aesthetics after repair. For areas with abnormal bonding layers, glue will be reapplied and pressed under pressure to make the bonding layer more firm.

[0106] In this embodiment, the defect evaluation index Qpzs comprehensively considers the flatness and burr distribution of the particleboard surface, providing a scientific basis for quality evaluation. Evaluation unit: According to the pre-set evaluation threshold Q, the defect evaluation index Qpzs is compared and analyzed to generate the corresponding quality status result. According to the evaluation results, the particleboard is automatically sorted into different areas: unqualified sorting area, repair sorting area and qualified sorting area. Processing of the repair sorting area: For the particleboard to be repaired, manual inspection is carried out, and the burrs are finely polished with appropriate tools such as sandpaper or grinder to ensure a smooth surface. For the sunken area, wood filler or resin is used for repair to ensure the aesthetics and quality after repair. For areas with abnormal bonding layer, glue is re-applied and pressed to ensure that the bonding quality meets the standard. Through the intelligent sorting and repair process, the system can identify and handle quality problems in a timely manner, further reduce the possibility of defective products entering the market, and improve the quality stability of the production line and the overall quality level of the product. Effective quality control and repair mechanisms help reduce scrap rates and waste in the production process, while improving resource utilization and production efficiency, with significant economic and environmental benefits for enterprises.

[0107] Example 7

[0108] Please refer to Figure 1 and Figure 2 Specifically: A particleboard surface quality visual inspection method comprises the following steps:

[0109] Step 1: Place the particleboard to be tested on a workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particleboard to be tested;

[0110] Step 2: Then use multiple sets of visual inspection instruments to capture the image data of each board surface of the particleboard to be inspected, and pre-process several sets of board surface image data to remove noise, smooth the image, and perform geometric correction. Then, perform feature extraction on the pre-processed image data to obtain relevant flatness data and relevant burr state data on each board surface, so as to construct a visual inspection data set;

[0111] Step 3: Then, based on the relevant burr status data, the burr distribution coefficient Mfxs of the particleboard is constructed. If the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued;

[0112] Step 4. After receiving the early warning instruction, the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb are extracted according to the relevant flatness data on each board surface. The bonding factor Njyz is obtained by correlating the bubble density Qpmd with the glue water content Jhz. The flatness coefficient Pzxs is obtained by correlating the bonding factor Njyz with the hole area ratio Dcmb. Combined with the trained quality prediction model, the defect assessment index Qpzs is obtained by fitting.

[0113] Step 5. Finally, the evaluation threshold Q is set, and compared with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface and perform screening operations.

[0114] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A particleboard surface quality visual inspection system, characterized by: It includes position locking module, visual inspection module, burr inspection module, plate quality analysis module and feedback module; The position locking module is used to place the particle board to be tested on the workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particle board to be tested; The visual inspection module is used to capture the image data of each board surface of the particleboard to be inspected by using multiple sets of visual inspection instruments, and preprocess the image data of several sets of board surfaces to remove noise, smooth the image, and perform geometric correction, and then perform feature extraction on the preprocessed image data to respectively obtain relevant flatness data and relevant burr state data on each board surface to construct a visual inspection data set; The burr detection module is used to construct the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, and if the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued externally; The board quality analysis module is used to extract the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb according to the relevant flatness data on each board surface after receiving the early warning instruction, and obtain the bonding factor Njyz by correlating the bubble density Qpmd with the glue water content Jhz, and obtain the flatness coefficient Pzxs by correlating the bonding factor Njyz with the hole area ratio Dcmb, and obtain the defect evaluation index Qpzs by fitting in combination with the trained quality prediction model; The feedback module is used to pre-set the evaluation threshold Q, and compare and analyze it with the defect evaluation index Qpzs, so as to comprehensively analyze the current state of the particleboard surface and perform screening operations.

2. A particleboard surface quality visual inspection system according to claim 1, characterized in that: The position locking module includes a plate position locking unit; The plate locking unit is used to install a fixing device on the workbench, place the particleboard to be tested on the workbench through the fixing device, and use a laser rangefinder to measure the height of the particleboard relative to the workbench, and obtain the position coordinates of the particleboard to be tested by establishing a three-dimensional coordinate system, wherein the origin of the three-dimensional coordinate system is set at the exact center position of the particleboard to be tested.

3. A particleboard surface quality visual inspection system according to claim 1, characterized in that: The visual detection module includes an image acquisition unit, a preprocessing unit and a feature extraction unit; The image acquisition unit is used to install several groups of sensors and several groups of visual inspection instruments around the workbench, and after the particle board position is locked, start the several groups of visual inspection instruments to capture the image data of each board surface of the particle board to be inspected; The pre-processing unit is used to convert the image of each board surface into a grayscale image and use a median filter to remove noise in the image; The feature extraction unit is used to extract key feature information in the image through image enhancement technology, identify defective parts in the image, and respectively obtain relevant flatness data and relevant burr state data on each board surface, wherein the relevant flatness data includes the bubble density Qpmd on each surface of the particle board, the glue water content Jhz in the particle board production process, the glue curing time Jgsc in the particle board production process, the hole area ratio Dcmb and the maximum hole depth Kdsd; The relevant burr status data includes the burr height Mcgd on the particleboard surface, and combined with statistical algorithms, the standard deviation of the burr height Mcgd is calculated. std and the mean value of burr height Mcgd avg .

4. A particleboard surface quality visual inspection system according to claim 3, characterized in that: The burr detection module includes a preliminary analysis unit and a preliminary warning unit; The preliminary analysis unit is used to analyze and calculate the burr distribution coefficient Mfxs of the particleboard according to the relevant burr state data, which is specifically obtained according to the following formula: The significance of this formula is to: evaluate the discrete degree of burr height on the particleboard surface; Among them, Mcgd std Expressed as the standard deviation of burr height, Mcgd avg Expressed as the mean burr height.

5. A particleboard surface quality visual inspection system according to claim 4, characterized in that: The preliminary warning unit is used to pre-set the distribution threshold and compare it with the burr distribution coefficient Mfxs of the particle board to preliminarily analyze the current burr state of the particle board surface. The specific analysis content is as follows: If the burr distribution coefficient Mfxs of the particleboard exceeds the distribution threshold, it will be preliminarily judged that the burrs on the particleboard surface are in an abnormal state and an early warning instruction will be issued; If the burr distribution coefficient Mfxs of the particleboard does not exceed the distribution threshold, it will be preliminarily determined that the burrs on the surface of the particleboard are not in an abnormal state.

6. A particleboard surface quality visual inspection system according to claim 3, characterized in that: The plate quality analysis module includes an interlayer penetration analysis unit, a surface analysis unit and a comprehensive analysis unit; The interlayer penetration analysis unit is used to obtain the bonding factor Njyz after linear normalization of the bubble density Qpmd and the glue water content Jhz in the relevant flattening data on each board surface, which is specifically obtained according to the following formula: Where n is the number of particleboard surfaces, i = 1, 2, 3, ..., n, Qpmd i is expressed as the bubble density of the ith plate surface, It is represented by the average density of bubbles on the particleboard surface, Jgsc is represented by the glue curing time, and a1, a2 and a3 are all weight values.

7. A particleboard surface quality visual inspection system according to claim 6, characterized in that: The surface analysis unit is used to calculate the flattening coefficient Pzxs according to the relevant flattening data on each board surface and the bonding factor Njyz after linear normalization processing, which is specifically obtained according to the following formula: Where Dcmb represents the proportion of hole area, Kdsd represents the maximum hole depth, C represents the first correction constant, and α and β are weight values.

8. A particleboard surface quality visual inspection system according to claim 7, characterized in that: The comprehensive analysis unit is used to transmit the flattening coefficient Pzxs and the burr distribution coefficient Mfxs to the quality prediction model based on the visual inspection data set, and obtain the defect evaluation index Qpzs after linear normalization processing, which is specifically obtained according to the following formula: Wherein, R represents the second correction constant, and F1 and F2 are weight values.

9. The particleboard surface quality visual inspection system according to claim 1, characterized in that: The feedback module includes an evaluation unit and a screening unit; The evaluation unit is used to compare and analyze the evaluation threshold Q with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface. The specific comparison content is as follows: If the defect evaluation index Qpzs is greater than the evaluation threshold Q, it indicates that the particleboard currently being tested is in an unqualified state, and a first result is generated at this time; If the defect evaluation index Qpzs is equal to the evaluation threshold Q, it means that the particleboard currently being tested is in an unqualified state, and a second result is generated at this time; If the defect evaluation index Qpzs is less than the evaluation threshold Q, it indicates that the particleboard currently being tested is in a qualified state, and a third result is generated at this time; The screening unit is used to screen the particleboards corresponding to the first result, the second result and the third result, automatically screening the particleboards corresponding to the first result to the unqualified sorting area, automatically screening the particleboards corresponding to the second result to the repair sorting area, and automatically screening the particleboards corresponding to the third result to the qualified sorting area; Among them, the particleboards in the sorting area to be repaired will be manually inspected, and the surface of the particleboard will be polished with sandpaper or a grinder to remove burrs. The filler should match the color and texture of the particleboard. For areas with abnormal bonding layers, glue will be re-applied and pressed under pressure.

10. A particleboard surface quality visual inspection method, used to implement a particleboard surface quality visual inspection system as claimed in any one of claims 1 to 9, characterized in that: The following steps are included: Step 1: Place the particleboard to be tested on a workbench in advance and establish a three-dimensional coordinate system to lock the position coordinates of the particleboard to be tested; Step 2: Then use multiple sets of visual inspection instruments to capture the image data of each board surface of the particleboard to be inspected, and pre-process several sets of board surface image data to remove noise, smooth the image, and perform geometric correction. Then, perform feature extraction on the pre-processed image data to obtain relevant flatness data and relevant burr state data on each board surface, so as to construct a visual inspection data set; Step 3: Then, based on the relevant burr status data, the burr distribution coefficient Mfxs of the particleboard is constructed. If the burr distribution coefficient Mfxs exceeds the distribution threshold, an early warning instruction is issued; Step 4. After receiving the early warning instruction, the bubble density Qpmd, the glue water content Jhz and the hole area ratio Dcmb are extracted according to the relevant flatness data on each board surface. The bonding factor Njyz is obtained by correlating the bubble density Qpmd with the glue water content Jhz. The flatness coefficient Pzxs is obtained by correlating the bonding factor Njyz with the hole area ratio Dcmb. Combined with the trained quality prediction model, the defect assessment index Qpzs is obtained by fitting. Step 5. Finally, the evaluation threshold Q is set, and compared with the defect evaluation index Qpzs to comprehensively analyze the current state of the particleboard surface and perform screening operations.

Citation Information

Cited By

  • Intelligent production method and system for fabricated wall

    CN118906179A

  • Intelligent production method and system for fabricated wall

    CN118906179B

  • Shaving board quality detection method and system based on image recognition

    CN119379701A

  • Appearance quality inspection method for plastic Christmas tree production

    CN119904457A

  • A method for inspecting the appearance of plastic Christmas trees

    CN119904457B