A packaging box sealing quality inspection system based on machine vision

Abstract

The present invention belongs to the technical field of packaging box production inspection, specifically a packaging box sealing quality inspection system based on machine vision, including a machine vision scanning module, an image processing and extraction module, a sealing quality inspection module, a grasping robot module and an interactive display module; the present invention realizes the automatic inspection of the sealing quality of the packaging box through machine vision technology, improves the inspection efficiency and reduces the enterprise inspection cost, avoids the quality problems caused by human misjudgment, and reasonably evaluates the control performance of the sealing performance and the inspection process through the early warning mechanism management and control module and timely warns, so as to remind the operator to conduct cause investigation and analysis and quickly make reasonable response and improvement measures, so as to ensure the efficient and smooth progress of the sealing and its inspection process, and improve the quality of the processed packaging box, with high intelligence and automation levels.

Description

Technical Field

[0001] The present invention relates to the technical field of packaging box production detection, and in particular to a packaging box sealing quality detection system based on machine vision. Background Art

[0002] The packaging box is a box used to package products. It plays a very important role in protecting the goods during the circulation process. It also bears the functions of beautifying the goods, conveying the product information, and facilitating consumers to carry and use. In the production process of packaging boxes, the sealing link is very important. The quality of the sealing directly affects the sealing, durability and aesthetics of the packaging box.

[0003] After the packaging box is sealed, its sealing quality needs to be tested. At present, it is difficult to realize automatic and accurate testing of the sealing quality of packaging boxes and eliminate unqualified products based on machine vision technology. It is also impossible to reasonably evaluate the sealing performance and the control performance of the testing process and provide timely warnings, which is not conducive to ensuring the efficient and smooth progress of the sealing and testing process. The level of intelligence and automation is low;

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a packaging box sealing quality inspection system based on machine vision, which solves the problems that the prior art is difficult to realize automatic and accurate inspection of the packaging box sealing quality based on machine vision technology and to eliminate unqualified products, and is unable to reasonably evaluate the sealing performance and the control performance of the inspection process and provide timely warnings, and has a low level of intelligence and automation.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A packaging box sealing quality inspection system based on machine vision includes a machine vision scanning module, an image processing and extraction module, a sealing quality inspection module, a grabbing manipulator module and an interactive display module; the machine vision scanning module is composed of an industrial camera and a light source device, and the industrial camera collects images of the sealing part of the packaging box with the assistance of the light source device, and sends the collected packaging box sealing image to the image processing and extraction module;

[0008] The image processing and extraction module performs preprocessing including denoising and contrast enhancement on the packaging box sealing image, and uses image segmentation technology to extract feature information of the sealing part, including edge contour and color distribution information, and sends the extracted feature information to the sealing quality detection module;

[0009] The sealing quality detection module analyzes and compares the extracted feature information according to the preset sealing quality standard, identifies the sealing defects and determines whether the sealing quality is qualified, and sends the judgment result to the interactive display module and the grasping robot module; when it is judged that the sealing quality of the package box is unqualified, the package box is sorted by the grasping robot module to remove the corresponding package box, and when it is judged that the sealing quality of the package box is qualified, it enters the next process.

[0010] Furthermore, the interactive display module is communicated with the early warning mechanism control module. The early warning mechanism control module is used to set the detection period, obtain the branding assessment symbol and the inspection assessment symbol of the detection period, and if FY-2∩TW-2 is obtained, the early warning mechanism will not be triggered. In other cases, the early warning mechanism will be triggered to generate corresponding early warning information, and the generated early warning information will be sent to the interactive display module. The interactive display module displays the corresponding early warning information and emits an alarm sound.

[0011] Furthermore, the early warning mechanism control module is in communication connection with the branding performance evaluation module and the operation control analysis evaluation module, and the branding performance evaluation module assigns a branding evaluation symbol FY-1 or FY-2 through analysis, and sends the branding evaluation symbol FY-1 or FY-2 to the early warning mechanism control module;

[0012] The operation control analysis and evaluation module assigns the inspection evaluation symbol TW-1 or TW-2 through analysis, and sends the inspection evaluation symbol TW-1 or TW-2 to the early warning mechanism control module.

[0013] Furthermore, the specific analysis process of the branding performance evaluation module is as follows:

[0014] Obtain the results of the seaming quality judgment of all packaging boxes produced within the detection period, mark the packaging boxes with unqualified seaming quality as unusable packaging boxes, calculate the ratio of the number of unusable packaging boxes to the total number of packaging boxes produced within the detection period to obtain the packaging box rejection coefficient; compare the packaging box rejection coefficient with the preset packaging box rejection coefficient threshold, and if the packaging box rejection coefficient exceeds the preset packaging box rejection coefficient threshold, assign the seaming evaluation symbol FY-1;

[0015] If the rejection coefficient of the packaging box does not exceed the preset rejection coefficient threshold of the packaging box, the defect type to be detected is obtained, and the corresponding defect type is marked as the target type i, where i is a natural number greater than 1; based on the sealing quality judgment result, all defects existing on the corresponding packaging box are obtained, and if the defects existing on the corresponding packaging box include the target type i, the corresponding packaging box is marked as a match for the target type i;

[0016] The number of matching products corresponding to target type i in the detection period is obtained and the ratio is calculated with the total number of packaging boxes produced in the detection period to obtain the matching product detection value, and the matching product detection value is numerically compared with the corresponding preset matching product detection threshold. If the matching product detection value exceeds the corresponding preset matching product detection threshold, the target type i is marked as a hidden danger type; if the hidden danger type exists in the detection period, the branding assessment symbol FY-1 is assigned.

[0017] Furthermore, if there is a hidden danger type within the detection period, the ratio of the matching product detection value of the target type i to the corresponding preset matching product detection threshold is marked as the matching risk value, the average matching risk value of all types of defects appearing within the detection period is marked as the seam risk table value, and the matching risk value with the largest value is marked as the seam risk amplitude value;

[0018] The brand-sealing quality inspection coefficient is obtained by numerically calculating the packaging box rejection coefficient, the brand-sealing risk table value and the brand-sealing risk amplitude value, and the brand-sealing quality inspection coefficient is numerically compared with the preset brand-sealing quality inspection coefficient threshold value. If the brand-sealing quality inspection coefficient exceeds the preset brand-sealing quality inspection coefficient threshold value, the brand-sealing assessment symbol FY-1 is assigned; if the brand-sealing quality inspection coefficient does not exceed the preset brand-sealing quality inspection coefficient threshold value, the brand-sealing assessment symbol FY-2 is assigned.

[0019] Furthermore, the specific analysis process of the operation control analysis and evaluation module is as follows:

[0020] Obtain the light source information of the area where the machine vision scanning module is located, including light brightness, light color and light color temperature, and evaluate whether the lighting environment of the area where the machine vision scanning module is located meets the requirements based on the light source information. If the lighting environment does not meet the requirements, it is determined that the corresponding area is in an abnormal lighting state;

[0021] The total duration of the abnormal lighting state in the corresponding area during the detection period is collected and marked as the abnormal lighting time value, and the number of occurrences in which the single duration of the abnormal lighting state in the corresponding area during the detection period exceeds the corresponding preset single duration threshold is marked as the light control risk value; the light anomaly value and the light control risk value are numerically compared with the preset light anomaly threshold and the preset light control risk threshold respectively; if the light anomaly value or the light control risk value exceeds the corresponding preset threshold, the inspection evaluation symbol TW-1 is assigned.

[0022] Furthermore, if both the illumination anomaly value and the illumination risk control value do not exceed the corresponding preset thresholds, the average value of the reaction time of the grasping manipulator module during the detection period is collected and marked as the grasping reaction coefficient, and the actual operation trajectory of each grasping operation of the grasping manipulator module during the detection period is collected, and the actual operation trajectory is overlapped and compared with the corresponding standard operation trajectory, thereby obtaining the proportion of the non-overlapping trajectory length in the corresponding grasping operation and marking it as the trajectory anomaly coefficient. If the trajectory anomaly coefficient exceeds the preset trajectory anomaly coefficient threshold, the corresponding grasping operation is marked as a risky operation;

[0023] The number of risky operations within the detection period is obtained and marked as the risk operation inspection value, and the average of all trajectory abnormality coefficients within the detection period is calculated to obtain the grasping track inspection value; the grasping control evaluation value is obtained by numerically calculating the grasping reaction coefficient, the risk operation inspection value and the grasping track inspection value, and the grasping control evaluation value is numerically compared with the preset grasping control evaluation threshold. If the grasping control evaluation value exceeds the preset grasping control evaluation threshold, the inspection evaluation symbol TW-1 is assigned; if the grasping control evaluation value does not exceed the preset grasping control evaluation threshold, the inspection evaluation symbol TW-2 is assigned.

[0024] Furthermore, the interactive display module is communicatively connected to the branding management analysis module, which is used to set a management period, analyze the difficulty of branding management of the packaging box within the management period, and determine whether to generate a branding management enhancement signal through analysis. When the branding management enhancement signal is generated, it is sent to the interactive display module for display.

[0025] Furthermore, the specific analysis process of the branding management analysis module is as follows:

[0026] The number of times the early warning mechanism is triggered by the early warning mechanism control module during the management period is collected and marked as the early warning mechanism trigger value, and the early warning mechanism trigger value is numerically compared with the preset early warning mechanism trigger threshold. If the early warning mechanism trigger value exceeds the preset early warning mechanism trigger threshold, a branding management enhancement signal is generated;

[0027] If the trigger value of the early warning mechanism does not exceed the preset trigger threshold of the early warning mechanism, the sealing temperature, sealing pressure and sealing speed of the sealing device during the sealing are collected, and the deviation value of the sealing temperature compared to the set standard sealing temperature is marked as the sealing temperature value. Similarly, the sealing pressure value and the sealing speed value are obtained;

[0028] The sealing operation value is calculated by weighted summing the sealing temperature value, the sealing pressure value and the sealing speed value, and the sealing operation value is numerically compared with the preset sealing operation threshold value. If the sealing operation value exceeds the preset sealing operation threshold value, the corresponding sealing operation value is marked as a sealing abnormal value;

[0029] The number of sealing anomaly table values ​​during the management period is obtained and the ratio is calculated with the number of sealing operation values ​​to obtain the sealing anomaly percentage value, and the sealing anomaly percentage value is numerically compared with the preset sealing anomaly percentage threshold value. If the sealing anomaly percentage value exceeds the preset sealing anomaly percentage threshold, a sealing management enhancement signal is generated.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. In the present invention, the image of the sealing part of the packaging box is collected by the machine vision scanning module, the image processing and extraction module pre-processes the sealing image of the packaging box and extracts the characteristic information of the sealing part, and the sealing quality detection module analyzes and compares the extracted characteristic information according to the preset sealing quality standard, identifies the defects of the sealing and judges whether the sealing quality is qualified. When the sealing quality of the packaging box is judged to be unqualified, the packaging box is sorted by the grasping manipulator module to reject the corresponding packaging box, thereby realizing the automatic detection of the sealing quality of the packaging box, improving the detection efficiency and the accuracy of the detection results and reducing the detection cost of the enterprise;

[0032] 2. In the present invention, the early warning mechanism control module is used to reasonably evaluate the control performance of the sealing performance and the detection process and issue a timely warning, so as to remind the operating personnel to conduct cause investigation and analysis and quickly make reasonable response and improvement measures to ensure the efficient and smooth progress of the sealing and its detection process and improve the quality of the processed packaging boxes. The sealing management and analysis module is used to reasonably evaluate the operational performance of the packaging box sealing and sealing quality detection during the management period to remind the operating personnel to strengthen subsequent supervision in a timely manner, ensure the subsequent operational stability and efficiency and improve the quality of the produced packaging boxes. The degree of intelligence is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0034] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0035] Figure 2 This is a system block diagram of Embodiment 2, Embodiment 3 and Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0036] 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.

[0037] Embodiment 1: Figure 1As shown, the present invention proposes a packaging box sealing quality inspection system based on machine vision, including a machine vision scanning module, an image processing and extraction module, a sealing quality inspection module, a grabbing manipulator module, an early warning mechanism control module and an interactive display module;

[0038] The machine vision scanning module is composed of an industrial camera and a light source device. The industrial camera is used for image acquisition, and the light source device is used for providing appropriate lighting. The industrial camera acquires images of the sealing part of the packaging box with the assistance of the light source device, and sends the acquired packaging box sealing images to the image processing and extraction module.

[0039] The image processing and extraction module pre-processes the packaging box seal image, including denoising and contrast enhancement to improve the image quality, and uses image segmentation technology to extract the characteristic information of the seal part, including edge contour and color distribution information, and sends the extracted characteristic information to the seal quality detection module to provide information support for the seal quality detection and analysis;

[0040] The sealing quality detection module analyzes and compares the extracted feature information according to the preset sealing quality standard, identifies the sealing defects (such as missing seals, misalignment, burning, etc.) and determines whether the sealing quality is qualified, and sends the judgment result to the interactive display module and the grasping robot module; when it is judged that the sealing quality of the package box is unqualified, the package box is sorted by the grasping robot module to remove the corresponding package box, and when it is judged that the sealing quality of the package box is qualified, it enters the next process.

[0041] The present invention realizes automated detection of the sealing quality of packaging boxes through machine vision technology, greatly improves detection efficiency, reduces the detection cost of enterprises, and can accurately identify sealing defects, avoid quality problems caused by human misjudgment, improve product quality, and help operators to formulate targeted response plans to ensure the sealing quality of packaging boxes.

[0042] The early warning mechanism control module is used to set the detection period, preferably, the detection period is sixty minutes; obtain the seal assessment symbol (FY-1 or FY-2) and the table inspection assessment symbol (TW-1 or TW-2) of the detection period, if FY-2∩TW-2 is obtained, the early warning mechanism is not triggered; in other cases (including FY-1∩TW-2, FY-2∩TW-1 or FY-1∩TW-1), the early warning mechanism is triggered to generate corresponding early warning information, and the generated early warning information is sent to the interactive display module;

[0043] The interactive display module displays the corresponding warning information and emits an alarm sound. It can make a reasonable assessment of the control performance of the sealing and detection process and issue a timely warning to remind the operator to conduct a cause investigation and analysis and quickly make reasonable response and improvement measures, ensuring the efficient and smooth progress of the sealing and detection process, improving the quality of the processed packaging boxes, and having a high level of intelligence and automation.

[0044] Embodiment 2: Figure 2 As shown, the difference between this embodiment and the first embodiment is that the early warning mechanism control module is connected to the sealing performance evaluation module in communication. The sealing performance evaluation module assigns the sealing evaluation symbol FY-1 or FY-2 through analysis, and sends the sealing evaluation symbol FY-1 or FY-2 to the early warning mechanism control module, which can not only accurately evaluate the sealing operation performance of the packaging box within the detection period, but also provide information support for the analysis process of the early warning mechanism control module to ensure the accuracy of its analysis and early warning results; the specific analysis process of the sealing performance evaluation module is as follows:

[0045] Obtain the results of the seaming quality judgment of all packaging boxes produced during the testing period, mark the packaging boxes with unqualified seaming quality as unusable packaging boxes, calculate the ratio of the number of unusable packaging boxes to the total number of packaging boxes produced during the testing period to obtain the packaging box rejection coefficient; compare the packaging box rejection coefficient with the preset packaging box rejection coefficient threshold, if the packaging box rejection coefficient exceeds the preset packaging box rejection coefficient threshold, indicating that the seaming quality performance of the packaging box during the testing period is poor, then assign the seaming evaluation symbol FY-1;

[0046] If the rejection coefficient of the packaging box does not exceed the preset rejection coefficient threshold of the packaging box, the defect type to be detected is obtained, and the corresponding defect type is marked as the target type i, where i is a natural number greater than 1; based on the sealing quality judgment result, all defects existing on the corresponding packaging box are obtained, and if the defects existing on the corresponding packaging box include the target type i, the corresponding packaging box is marked as a match for the target type i;

[0047] The number of matching products corresponding to target type i in the detection period is obtained and the ratio is calculated with the total number of packaging boxes produced in the detection period to obtain a matching product detection value, and the matching product detection value is numerically compared with the corresponding preset matching product detection threshold. If the matching product detection value exceeds the corresponding preset matching product detection threshold, it indicates that the defects corresponding to target type i in the detection period occur frequently, and the resulting sealing quality risks are relatively large, and the target type i is marked as a potential risk type;

[0048] If there is a hidden danger type during the detection period, it means that there is a hidden danger of branding during the detection period, which is not conducive to ensuring the quality of branding for the packaging box, and the branding assessment symbol FY-1 is assigned.

[0049] Furthermore, if there is a hidden danger type in the detection period, the ratio of the matching product detection value of the target type i to the corresponding preset matching product detection threshold is marked as the matching risk value, the average matching risk value of all types of defects appearing in the detection period is marked as the seam sealing risk table value, and the matching risk value with the largest value is marked as the seam sealing risk amplitude value;

[0050] The package rejection coefficient SY, the sealing risk table value QS and the sealing risk amplitude value YF are numerically calculated by the formula YM=k×SY+(a×QS+c×YF) / 2 to obtain the sealing quality inspection coefficient YM; wherein k, a, c are preset proportional coefficients, and k>a>c>0; and the larger the value of the sealing quality inspection coefficient YM, the worse the sealing quality performance of the package box is in general during the detection period;

[0051] The sealing quality inspection coefficient YM is numerically compared with the preset sealing quality inspection coefficient threshold. If the sealing quality inspection coefficient YM exceeds the preset sealing quality inspection coefficient threshold, indicating that the sealing quality performance of the packaging box during the detection period is generally poor, the sealing evaluation symbol FY-1 is assigned; if the sealing quality inspection coefficient YM does not exceed the preset sealing quality inspection coefficient threshold, indicating that the sealing quality performance of the packaging box during the detection period is generally good, the sealing evaluation symbol FY-2 is assigned.

[0052] Embodiment 3: Figure 2 As shown, the difference between this embodiment and the first and second embodiments is that the early warning mechanism control module is connected to the operation control analysis and evaluation module in communication. The operation control analysis and evaluation module assigns the table inspection evaluation symbol TW-1 or TW-2 through analysis, and sends the table inspection evaluation symbol TW-1 or TW-2 to the early warning mechanism control module, which can not only accurately evaluate the control performance of the sealing quality inspection process during the inspection period, but also provide information support for the analysis process of the early warning mechanism control module, and further ensure the accuracy of its analysis and early warning results; the specific analysis process of the operation control analysis and evaluation module is as follows:

[0053] The light source information of the area where the machine vision scanning module is located is obtained, including light brightness, light color and light color temperature. Based on the light source information, whether the lighting environment of the area where the machine vision scanning module is located meets the requirements (i.e., whether the light brightness, light color and light color temperature meet the requirements) is evaluated. If the lighting environment does not meet the requirements, it is judged that the corresponding area is in an abnormal lighting state (i.e., the light brightness, light color or light color temperature does not meet the requirements); it should be noted that when the corresponding area is in an abnormal lighting state, it is not conducive to ensuring the quality of the collected image;

[0054] The total duration of the abnormal lighting state in the corresponding area during the detection period is collected and marked as the abnormal lighting time value, and the number of occurrences in which the single duration of the abnormal lighting state in the corresponding area during the detection period exceeds the corresponding preset single duration threshold is marked as the light control risk value; the light abnormal time value and the light control risk value are numerically compared with the preset light abnormal time threshold and the preset light control risk threshold respectively. If the light abnormal time value or the light control risk value exceeds the corresponding preset threshold, indicating that the control performance of the regional lighting during the detection period is poor, the inspection evaluation symbol TW-1 is assigned.

[0055] Furthermore, if both the illumination abnormality value and the illumination risk control value do not exceed the corresponding preset threshold value, the average value of the reaction time of the grasping robot module in the detection period is collected and marked as the grasping reaction coefficient; it should be noted that the reaction time of the grasping robot module is the interval between the moment when the sealing quality of the packaging box is judged to be unqualified and the moment when the grasping robot module starts to grasp; and the larger the value of the grasping reaction coefficient, the slower the grasping reaction of the grasping robot module;

[0056] And collect the actual operation trajectory of each grasping operation of the grasping manipulator module during the detection period, overlap and compare the actual operation trajectory with the corresponding standard operation trajectory, and obtain the proportion of the non-overlapping trajectory length in the corresponding grasping operation and mark it as the trajectory abnormality coefficient. If the trajectory abnormality coefficient exceeds the preset trajectory abnormality coefficient threshold, it indicates that the execution trajectory of the corresponding grasping operation is not standard, and the corresponding grasping operation is marked as a risky operation;

[0057] The number of risky operations in the detection period is obtained and marked as the risky operation inspection value, and the average of all trajectory abnormality coefficients in the detection period is calculated to obtain the grasping track inspection value;

[0058] The grasping reaction coefficient TN, the risk operation inspection value WF and the grasping track inspection value BT are numerically calculated by the formula ZP=fu×TN+mu×WF+hg×BT to obtain the grasping control evaluation value ZP; wherein fu, mu, and hg are preset proportional coefficients with values ​​greater than zero, and the larger the value of the grasping control evaluation value ZP, the worse the overall control performance of the grasping manipulator module during the detection period;

[0059] The grasping control evaluation value ZP is numerically compared with the preset grasping control evaluation threshold. If the grasping control evaluation value ZP exceeds the preset grasping control evaluation threshold, it indicates that the control performance of the grasping robot module during the detection period is generally poor, and the inspection evaluation symbol TW-1 is assigned; if the grasping control evaluation value ZP does not exceed the preset grasping control evaluation threshold, it indicates that the control performance of the grasping robot module during the detection period is generally good, and the inspection evaluation symbol TW-2 is assigned.

[0060] Embodiment 4: Figure 2As shown, the difference between this embodiment and the first, second and third embodiments is that the interactive display module is communicatively connected to the branding management and analysis module, the branding management and analysis module is communicatively connected to the early warning mechanism control module, the branding management and analysis module is used to set a management period, preferably, the management period is seven days; the branding management difficulty of the packaging box within the management period is analyzed, and the analysis is used to determine whether to generate a branding management enhancement signal;

[0061] When the sealing management enhancement signal is generated, it is sent to the interactive display module for display, which can reasonably evaluate the operation performance of the packaging box sealing and its sealing quality detection during the management period, so as to remind the operator to strengthen the subsequent supervision in time, ensure the subsequent operation stability and efficiency and improve the quality of the produced packaging boxes, with a high degree of intelligence; the specific analysis process of the sealing management analysis module is as follows:

[0062] The number of times the early warning mechanism is triggered by the early warning mechanism control module during the management period is collected and marked as the early warning mechanism trigger value, and the early warning mechanism trigger value is numerically compared with the preset early warning mechanism trigger threshold. If the early warning mechanism trigger value exceeds the preset early warning mechanism trigger threshold, it indicates that the management performance of the sealing quality detection operation during the management period is poor and the management difficulty is greater, then a sealing management enhancement signal is generated;

[0063] If the trigger value of the early warning mechanism does not exceed the preset trigger threshold of the early warning mechanism, the sealing temperature, sealing pressure and sealing speed of the sealing equipment during sealing are collected, and the deviation value of the sealing temperature compared with the set standard sealing temperature is marked as the sealing temperature condition value, the deviation value of the sealing pressure compared with the set standard sealing pressure is marked as the sealing pressure condition value, and the deviation value of the sealing speed compared with the set standard sealing speed is marked as the sealing speed condition value;

[0064] The sealing temperature condition value X, the sealing pressure condition value W and the sealing speed condition value R are weighted and calculated by the formula LF=e×X+t×W+n×R to obtain the sealing operation value LF, wherein e, t and n are preset weight coefficients whose values ​​are greater than zero, and the larger the value of the sealing operation value LF is, the more abnormal the real-time sealing operation condition of the sealing device is in general; the sealing operation value LF is numerically compared with the preset sealing operation threshold value, and if the sealing operation value LF exceeds the preset sealing operation threshold value, it indicates that the real-time sealing operation condition of the sealing device is in general poor, and the corresponding sealing operation value is marked as a sealing abnormal value;

[0065] The number of sealing abnormality table values ​​during the management period is obtained and the ratio is calculated with the number of sealing operation values ​​to obtain the sealing abnormality percentage value, and the sealing abnormality percentage value is numerically compared with the preset sealing abnormality threshold value. If the sealing abnormality percentage value exceeds the preset sealing abnormality threshold value, it indicates that the sealing operation performance of the sealing equipment during the management period is poor and the management of the sealing operation is difficult, then a sealing management enhancement signal is generated.

[0066] The working principle of the present invention is as follows: when in use, the image of the sealing part of the packaging box is collected by the machine vision scanning module, the image processing and extraction module pre-processes the sealing image of the packaging box and extracts the characteristic information of the sealing part, and the sealing quality detection module analyzes and compares the extracted characteristic information according to the preset sealing quality standard, identifies the defects of the sealing and judges whether the sealing quality is qualified, and when it is judged that the sealing quality of the packaging box is unqualified, the packaging box is sorted by the grasping robot module to remove the corresponding packaging box, thereby realizing the automatic detection of the sealing quality of the packaging box, improving the detection efficiency and reducing the detection cost of the enterprise, avoiding the quality problems caused by manual misjudgment, and through the early warning mechanism management and control module, the sealing performance and the control performance of the detection process are reasonably evaluated and warned in time, so as to remind the operator to conduct cause investigation and analysis and quickly make reasonable response and improvement measures, so as to ensure the efficient and smooth progress of the sealing and its detection process, improve the quality of the processed packaging box, and have a high level of intelligence and automation.

[0067] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the contents of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A packaging box sealing quality inspection system based on machine vision, characterized in that: It includes a machine vision scanning module, an image processing and extraction module, a sealing quality detection module, a grabbing manipulator module and an interactive display module; the machine vision scanning module is composed of an industrial camera and a light source device. The industrial camera collects images of the sealing parts of the packaging box with the assistance of the light source device, and sends the collected packaging box sealing images to the image processing and extraction module; The image processing and extraction module performs preprocessing including denoising and contrast enhancement on the packaging box sealing image, and uses image segmentation technology to extract feature information of the sealing part, and sends the extracted feature information to the sealing quality detection module; The sealing quality detection module analyzes and compares the extracted feature information according to the preset sealing quality standard, identifies the sealing defects and determines whether the sealing quality is qualified, and sends the judgment result to the interactive display module and the grasping robot module; when it is judged that the sealing quality of the package box is unqualified, the package box is sorted by the grasping robot module to remove the corresponding package box, and when it is judged that the sealing quality of the package box is qualified, it enters the next process.

2. The packaging box sealing quality inspection system based on machine vision according to claim 1 is characterized in that: The interactive display module is communicated with the early warning mechanism control module. The early warning mechanism control module is used to set the detection period, obtain the branding assessment symbol and the inspection assessment symbol of the detection period, and if FY-2∩TW-2 is obtained, the early warning mechanism will not be triggered. In other cases, the early warning mechanism will be triggered to generate corresponding early warning information, and the generated early warning information will be sent to the interactive display module. The interactive display module displays the corresponding early warning information and emits an alarm sound.

3. The packaging box sealing quality inspection system based on machine vision according to claim 2 is characterized in that: The early warning mechanism control module is connected to the branding performance evaluation module and the operation control analysis evaluation module in communication. The branding performance evaluation module assigns a branding evaluation symbol FY-1 or FY-2 through analysis, and sends the branding evaluation symbol FY-1 or FY-2 to the early warning mechanism control module. The operation control analysis and evaluation module assigns the inspection evaluation symbol TW-1 or TW-2 through analysis, and sends the inspection evaluation symbol TW-1 or TW-2 to the early warning mechanism control module.

4. The packaging box sealing quality inspection system based on machine vision according to claim 3 is characterized in that: The specific analysis process of the branding performance evaluation module is as follows: Obtain the results of the seaming quality judgment of all packaging boxes produced during the testing period, mark the packaging boxes with unqualified seaming quality as unusable packaging boxes, and calculate the ratio of the number of unusable packaging boxes to the total number of packaging boxes produced during the testing period to obtain the packaging box rejection coefficient; If the box rejection coefficient exceeds the preset box rejection coefficient threshold, the seal assessment symbol FY-1 is assigned; If the packaging box rejection coefficient does not exceed the preset packaging box rejection coefficient threshold, the defect type to be detected is obtained, and the corresponding defect type is marked as target type i, where i is a natural number greater than 1; Based on the seal quality judgment result, all defects existing on the corresponding packaging box are obtained. If the defects existing on the corresponding packaging box include target type i, the corresponding packaging box is marked as a matching product of target type i; The number of matching products corresponding to target type i in the detection period is obtained and the ratio is calculated with the total number of packaging boxes produced in the detection period to obtain the matching product detection value. If the matching product detection value exceeds the corresponding preset matching product detection threshold, the target type i is marked as a hidden danger type; if a hidden danger type exists in the detection period, the branding assessment symbol FY-1 is assigned.

5. The packaging box sealing quality inspection system based on machine vision according to claim 4 is characterized in that: If there is a hidden danger type within the detection period, the branding quality inspection coefficient is obtained by numerically calculating the packaging box rejection coefficient, the branding risk table value and the branding risk amplitude. If the branding quality inspection coefficient exceeds the preset branding quality inspection coefficient threshold, the branding assessment symbol FY-1 is assigned; if the branding quality inspection coefficient does not exceed the preset branding quality inspection coefficient threshold, the branding assessment symbol FY-2 is assigned.

6. The packaging box sealing quality inspection system based on machine vision according to claim 3 is characterized in that: The specific analysis process of the operation control analysis and evaluation module is as follows: Obtain the light source information of the area where the machine vision scanning module is located. If the lighting environment does not meet the requirements, it is determined that the corresponding area is in an abnormal lighting state; The total duration of the abnormal lighting state in the corresponding area during the detection period is collected and marked as the abnormal lighting time value, and the number of occurrences of the single duration of the abnormal lighting state in the corresponding area during the detection period exceeding the corresponding preset single duration threshold is marked as the light control risk value; if the abnormal lighting time value or the light control risk value exceeds the corresponding preset threshold, the inspection evaluation symbol TW-1 is assigned.

7. The packaging box sealing quality inspection system based on machine vision according to claim 6, characterized in that: If both the light variability value and the light control risk value do not exceed the corresponding preset thresholds, the grasping control evaluation value is obtained by numerically calculating the grasping reaction coefficient, the risk operation inspection value and the grasping track inspection value. If the grasping control evaluation value exceeds the preset grasping control evaluation threshold, the inspection evaluation symbol TW-1 is assigned; if the grasping control evaluation value does not exceed the preset grasping control evaluation threshold, the inspection evaluation symbol TW-2 is assigned.

8. The packaging box sealing quality inspection system based on machine vision according to claim 2, characterized in that: The interactive display module is communicatively connected to the branding management analysis module. The branding management analysis module is used to set a management period, analyze the difficulty of branding management of the packaging box during the management period, and determine whether to generate a branding management enhancement signal through analysis. When the branding management enhancement signal is generated, it is sent to the interactive display module for display.

9. The packaging box sealing quality inspection system based on machine vision according to claim 8, characterized in that: The specific analysis process of the branding management analysis module is as follows: The number of times the early warning mechanism control module triggers the early warning mechanism during the management period is collected and marked as the early warning mechanism trigger value. If the early warning mechanism trigger value exceeds the preset early warning mechanism trigger threshold, a branding management enhancement signal is generated; If the early warning mechanism trigger value does not exceed the preset early warning mechanism trigger threshold, the sealing operation value is calculated by weighted summing up the sealing temperature value, the sealing pressure value and the sealing speed value; if the sealing operation value exceeds the preset sealing operation threshold, the corresponding sealing operation value is marked as a sealing anomaly value; the number of sealing anomaly values ​​during the management period is obtained and the ratio is calculated by comparing it with the number of sealing operation values ​​to obtain the sealing anomaly percentage value; if the sealing anomaly percentage value exceeds the preset sealing anomaly percentage threshold, a sealing management enhancement signal is generated.

Images