Visual inspection system and method applied to bag making machine

Abstract

The invention discloses a visual inspection system and method applied to a bag making machine, and belongs to the technical field of visual inspection, and the method comprises the following steps: establishing a database, storing front and back image information and a two-dimensional coordinate graph according to numbers, shooting a complete image of two layers of films, and when the complete image is received for the first time, carrying out visual inspection on the two layers of films; searching the same front and back image information in a database according to the front image information and the back image information, if the same front and back image information cannot be found, giving an alarm, comparing a distinguishing region, judging whether the distinguishing region is fuzzy or blank, calculating coordinates of a fuzzy region and a blank region, and judging whether the distinguishing region is fuzzy or blank; the coordinate difference of the blank contrast image information is calculated, if the same front and back image information is found, the front and back image information serves as the closest front and back image information, and the method has the effects that the film printing quality can be automatically detected, and the possible error reason can be automatically judged.

Description

Technical Field

[0001] The present invention relates to the field of visual inspection, and in particular to a visual inspection system and method applied to a bag making machine. Background Art

[0002] At present, a bag making machine processes a roll of film that has been processed through printing, lamination, slitting, cutting and other processes. Under the action of a bag-shaped mold, it is heated, sealed, cut and then rewound to make bags. The printed film will be divided into two layers, front and back, on the bag making machine. The two layers of film are laminated by heat sealing or other methods and then cut into bags. During the processing, it is usually necessary to conduct a qualified inspection on the product to determine whether there are defective products. For bags, possible problems include but are not limited to bag body damage, printing errors, and film misalignment. Among them, printing errors are the most common, especially when changing the products to be produced.

[0003] The above-mentioned existing technical solutions have the following defects: Although the bag making machine can detect printing pattern errors in a timely manner, it still requires manual full-process investigation to determine the cause of the error, which is time-consuming and laborious. Summary of the Invention

[0004] In order to conveniently and simply monitor the printing quality of the film on the bag making machine, the present application provides a visual inspection system and method applied to a bag making machine.

[0005] On the one hand, a visual inspection method applied to a bag making machine provided by the present application adopts the following technical solutions: A visual inspection method applied to a bag making machine further includes the following steps: Establish a database, store front and back image information and two-dimensional coordinate maps according to numbers, and establish coordinates for each piece of front and back image information based on the two-dimensional coordinate map; Take a complete image of the two layers of film to obtain front image information and back image information; When receiving the complete image for the first time, search for the same front and back image information in the database according to the front image information and the back image information; If the same front and back image information cannot be found, an alarm is issued and the closest front and back image information is searched in the database. Compare the difference area between the received complete image and the closest front and back image information and frame the difference area; Judge whether the image in the framed difference area is blurred, and at the same time judge whether the image in the framed difference area is blank; Mark the framed difference area determined to be blurred as a blurred area, mark the blurred area on the closest front and back image information to obtain blurred comparison image information, calculate the coordinates of the blurred area, display the blurred comparison image information and display the coordinates; Mark the selected difference area judged as blank as a blank area, mark the blank area on the closest front and back image information to obtain blank comparison image information, calculate the coordinates of the image edge and the blank area in the blank comparison image information, and compare them with the corresponding coordinates of the closest front and back image information. If the coordinates of the image edges of the two are different, calculate the coordinate difference and display it. If the coordinates of the image edges of the two are the same, display the blank comparison image information and the coordinates of the blank area; If the same front and back image information is found, use this front and back image information as the closest front and back image information until the complete image is completely different from the found front and back image information.

[0006] By adopting the above solution, during the working process of the bag-making machine, this system can automatically take pictures of the front and back sides of the film, and can automatically determine the pattern and corresponding number of the current product through the database. If a printing error is found, it will automatically judge the closest product, and judge whether the printing is blurred or blank according to the pattern comparison of the product itself. If the printing is blurred, mark the blurred position. If there is a blank, it will first judge whether the whole-page pattern is printed out of position. If there is a printing misalignment, display the misalignment amount, otherwise display the marked blank position, which is convenient for users to judge printing faults. Users can know at any time through the system whether there is a printing error in the bag-making machine, and whether the printing error is pattern blurring, partial area blank or printing layout misalignment.

[0007] Preferably, it further includes the following steps: Set the number of emergency stop triggers; If the number of consecutive alarms reaches the number of emergency stop triggers, control the bag-making machine to stop. If a check instruction is received, control the bag-making machine to move the film until the same front and back image information is found, and then control the bag-making machine to work normally.

[0008] By adopting the above solution, after consecutive multiple alarms are triggered, it can be confirmed that the entire roll of film has problems. At this time, the system automatically controls the bag-making machine to stop, avoiding greater losses caused by continued processing. The staff can also manually input a detection instruction to make the bag-making machine only move the film. After determining that the pattern is intact, the processing ability of the bag-making machine will be automatically started to continue normal production.

[0009] Preferably, the step of "judging whether the image in the selected difference area is blurred" further includes: Set the blur judgment length; Identify the pixel cells of the image in the difference area and divide the pixel boundary of the image in the difference area; Judge the part of the image pixel boundary that is a line segment or arc as a regular boundary, and judge the remaining image pixel boundary as an irregular boundary; Judge whether the length of the irregular boundary exceeds the blur judgment length; If the length of the irregular boundary exceeds the fuzzy judgment length, the image within the difference region is determined to be blurred.

[0010] By adopting the above solution, after dividing the image into boundary lines according to pixel points, it is possible to determine whether the image is blurred by whether the boundary line is regular. A blurred image has almost no regular pixel boundaries, with accurate judgment and fast judgment speed.

[0011] Preferably, the step of "calculating the coordinates of the blurred region, displaying the blurred comparison image information and displaying the coordinates" further includes: Calling the coordinates of the front and back image information corresponding to the blurred region stored in the database, determining the 0-point coordinates through the same part of the blurred comparison image information and the front and back image information, and obtaining the coordinates of the boundary of the blurred region relative to the two-dimensional coordinate map after overlapping the blurred comparison image information with the front and back image information; Marking the coordinates of the boundary of the blurred region on the blurred comparison image information; Displaying the blurred comparison image information with the marked coordinates; The step of "calculating the coordinates of the image edge and the blank region in the blank comparison image information and comparing them with the corresponding coordinates of the closest front and back image information" further includes; Calling the coordinates of the front and back image information corresponding to the blank region stored in the database, determining the 0-point coordinates through the same part of the blank comparison image information and the front and back image information, and obtaining the coordinates of the boundary of the blank comparison image information and the blank region relative to the two-dimensional coordinate map after overlapping the blank comparison image information with the front and back image information; Comparing the boundary coordinates of the blank comparison image information with the boundary coordinates of the front and back image information.

[0012] By adopting the above solution, in combination with the front and back image information, the coordinates of any point on the blurred comparison image information and the blank comparison image information can be obtained, and the wrong position and whether the pattern is misaligned can be accurately judged through the coordinates.

[0013] Preferably, the following steps are further included: After confirming the closest front and back image information, when receiving the complete image, comparing the complete image with the determined front and back image information to determine whether the received complete image is the same as the determined front and back image information; If they are not the same, find the difference region between the received complete image and the determined front and back image information. If the received complete image is completely different from the determined front and back image information, an alarm is issued and the same front and back image information is searched again in the database.

[0014] By adopting the above solution, it is confirmed that the closest front and back image information can confirm the product corresponding to this roll of film, enabling the subsequent judgment process to be faster and consuming less computing power. If other patterns appear, it indicates that the product has changed, and the system can automatically switch the image information for judgment without manual operation by the user.

[0015] On the other hand, a vision detection system applied to a bag-making machine provided by the present application adopts the following technical solution: A vision detection system applied to a bag-making machine includes a data storage module, an image acquisition module, an image detection module, an error detection module, a fuzzy judgment module, a blank judgment module, and a fault display module; The data storage module stores the front and back image information and the two-dimensional coordinate map according to the number, and each front and back image information establishes coordinates based on the two-dimensional coordinate map; The image acquisition module includes two image shooting devices arranged on the bag-making machine. The two image shooting devices respectively shoot the complete images of the two layers of film, upload the front image information and the back image information, and transmit the front image information and the back image information to the image detection module; When the image detection module receives the complete image for the first time, it calls the front and back image information stored in the data storage module, searches for the same front and back image information according to the front image information and the back image information. If the same front and back image information cannot be found, it issues an alarm and transmits the error information to the error detection module. If the same front and back image information is found, the front and back image information is used as the closest front and back image information until the complete image is completely different from the found front and back image information; The error detection module calls the information stored in the data storage module, searches for the closest front and back image information according to the error information, compares and frames the difference area between the received complete image and the closest front and back image information, judges whether the image in the framed difference area is blurred, and at the same time judges whether the image in the framed difference area is blank. The framed difference area judged to be blurred is marked as a blurred area, and the blurred area is marked on the closest front and back image information to obtain the blurred comparison image information, and the blurred comparison image information is transmitted to the fuzzy judgment module; the framed difference area judged to be blank is marked as a blank area, and the blank area is marked on the closest front and back image information to obtain the blank comparison image information, and the blank comparison image information is transmitted to the blank judgment module; The fuzzy judgment module calculates the coordinates of the blurred area and transmits the blurred comparison image information with coordinates to the fault display module; The blank determination module calculates the coordinates of the image edges and blank areas in the blank comparison image information, and compares them with the corresponding coordinates of the closest front and back image information. If the coordinates of the image edges of the two are different, the coordinate difference is calculated, and the coordinate difference and the blank comparison image information are transmitted to the fault display module. If the coordinates of the image edges of the two are the same, the blank comparison image information with coordinates is transmitted to the fault display module; After receiving the image information, the fault display module performs a display.

[0016] By adopting the above solution, during the working process of the bag-making machine, this system can automatically capture the images of the front and back sides of the film. Through the database, it can automatically determine the pattern and corresponding number of the current product. If a printing error is found, it will automatically judge the closest product, and judge whether the printing is blurred or blank according to the pattern comparison of the product itself. If the printing is blurred, the blurred position will be marked. If there is a blank, it will first judge whether the entire pattern is printed out of position. If there is a printing misalignment, the misalignment amount will be displayed. Otherwise, the blank position will be marked, which is convenient for users to judge printing faults. Users can know at any time through the system whether there is a printing error in the bag-making machine, and whether the printing error belongs to pattern blurring, partial area blanking or printing layout misalignment.

[0017] Preferably, it further includes a fault emergency stop module. The fault emergency stop module receives the alarm from the image detection module. The fault emergency stop module presets the emergency stop trigger times. If the number of consecutive alarms reaches the emergency stop trigger times, it controls the bag-making machine to stop. If it receives an inspection instruction, it controls the bag-making machine to move the film until the same front and back image information is found, and then controls the bag-making machine to work normally.

[0018] By adopting the above solution, after the alarm is triggered continuously for multiple times, it can be confirmed that the entire roll of film has problems. At this time, the system automatically controls the bag-making machine to stop, avoiding greater losses caused by continued processing. The staff can also manually input a detection instruction to make the bag-making machine only move the film. After determining that the pattern is intact, the processing ability of the bag-making machine will be automatically started to continue normal production.

[0019] Preferably, the error detection module presets a fuzzy judgment length. The error detection module identifies the pixel cells of the image in the difference area, divides the image pixel boundaries in the difference area, determines the parts of the image pixel boundaries that are line segments or arcs as regular boundaries, and determines the remaining image pixel boundaries as irregular boundaries. It judges whether the length of the irregular boundary exceeds the fuzzy judgment length. If the length of the irregular boundary exceeds the fuzzy judgment length, it determines that the image in the difference area is blurred.

[0020] By adopting the above solution, after the image is divided into boundary lines according to pixel points, it can be judged whether the image is blurred by whether the boundary lines are regular. Blurred images hardly have regular pixel boundaries, and the judgment is accurate and fast.

[0021] Preferably, the fuzzy judgment module calls the coordinates of the front and back image information corresponding to the fuzzy area stored in the data storage module, determines the 0-point coordinates by fuzzy comparing the same parts of the image information and the front and back image information, obtains the coordinates of the boundary of the fuzzy area relative to the two-dimensional coordinate map after overlapping the fuzzy comparison image information with the front and back image information, marks the coordinates of the boundary of the fuzzy area on the fuzzy comparison image information, and transmits the marked fuzzy comparison image information to the fault display module; The blank judgment module calls the coordinates of the front and back image information corresponding to the blank area stored in the data storage module, determines the 0-point coordinates by blank comparing the same parts of the image information and the front and back image information, obtains the coordinates of the boundary of the blank comparison image information and the blank area relative to the two-dimensional coordinate map after overlapping the blank comparison image information with the front and back image information, and compares the boundary coordinates of the blank comparison image information with the boundary coordinates of the front and back image information.

[0022] By adopting the above scheme, the coordinates of any point on the fuzzy comparison image information and the blank comparison image information can be obtained in combination with the front and back image information, and the error position and whether the pattern is misaligned can be accurately judged through the coordinates.

[0023] Preferably, after the image detection module confirms the closest front and back image information, when receiving a complete image, it compares the complete image with the determined front and back image information to judge whether the received complete image is the same as the determined front and back image information. If they are not the same, it finds the difference area between the received complete image and the determined front and back image information. If the received complete image is completely different from the determined front and back image information, it issues an alarm and re-searches for the same front and back image information in the data storage module.

[0024] By adopting the above scheme, confirming the closest front and back image information can confirm the product corresponding to this roll of film, which can make the subsequent judgment process faster and consume less computing power. If there is another pattern, it means that the product has changed, and the system can automatically switch the image information for judgment without manual operation by the user.

[0025] In summary, the present invention has the following beneficial effects: 1. During the working process of the bag-making machine, this system can automatically take pictures of the front and back sides of the film, automatically determine the pattern and corresponding number of the current product through the database. If a printing error is found, it will automatically judge the closest product, and judge whether the printing is fuzzy or blank according to the pattern of the product itself. If the printing is fuzzy, it will mark the fuzzy position. If there is a blank, it will first judge whether the whole-page pattern is printed out of position. If there is a printing misalignment, it will display the misalignment amount, otherwise it will display and mark the blank position, which is convenient for users to judge printing faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is the overall system block diagram of the second embodiment of the present application.

[0027] Explanation of reference numerals in the drawings: 1. Data storage module; 2. Image acquisition module; 3. Image detection module; 4. Error detection module; 5. Fuzzy judgment module; 6. Blank judgment module; 7. Fault display module; 8. Fault emergency stop module. Specific implementation manners

[0028] The following further elaborates on the present application in conjunction with the attached Figure 1 drawings.

[0029] Embodiment 1. The present application discloses a vision detection method applied to a bag-making machine, and the specific steps are as follows: S100. Establish a database, and store the front and back image information and the two-dimensional coordinate map according to the number. Each piece of front and back image information is established with coordinates based on the two-dimensional coordinate map.

[0030] S101. Set the emergency stop trigger times and the fuzzy judgment length.

[0031] S200. Take a complete image of two layers of film to obtain the front image information and the back image information. An industrial camera can be set at the position where the film is stretched and unfolded before the heating step of the bag-making machine to photograph the outer side of the two layers of film.

[0032] S300. When receiving the complete image for the first time, search for the same front and back image information in the database according to the front image information and the back image information.

[0033] S301. If the same front and back image information cannot be found, an alarm is issued and the closest front and back image information is searched in the database. Compare the difference area between the received complete image and the closest front and back image information and frame the difference area.

[0034] S310. If the number of consecutive alarms reaches the emergency stop trigger times, control the bag-making machine to stop. If a check instruction is received, control the bag-making machine to move the film until the same front and back image information is found, and then control the bag-making machine to work normally. After consecutive alarms are triggered multiple times, it can be confirmed that the entire roll of film has problems. At this time, the system automatically controls the bag-making machine to stop to avoid greater losses caused by continued processing. The staff can also manually input a detection instruction to make the bag-making machine only move the film. After determining that the pattern is intact, the processing ability of the bag-making machine will be automatically started to continue normal production.

[0035] S302. Determine whether the image in the framed difference area is blurred: Identify the pixel cells of the image in the difference area and divide the pixel boundaries of the image in the difference area.

[0036] S303. Determine the part of the image pixel boundary that is a line segment or arc as a regular boundary, and determine the remaining image pixel boundary as an irregular boundary.

[0037] S304. Determine whether the length of the irregular boundary exceeds the fuzzy judgment length.

[0038] S305. If the length of the irregular boundary exceeds the fuzzy judgment length, determine that the image within the difference region is blurred.

[0039] S306. Determine whether the image within the boxed difference region is blank: When all pixels of the image within the difference region are the color of the back of the film (usually white or transparent, if transparent, set to the background image captured by the industrial camera without the film), then determine that the difference region is blank.

[0040] S400. Mark the boxed difference region determined to be blurred as a blurred region, and mark the blurred region on the closest front and back image information to obtain blurred comparison image information.

[0041] S401. Calculate the coordinates of the blurred region, display the blurred comparison image information and display the coordinates. Call the coordinates of the front and back image information corresponding to the blurred region stored in the database, determine the 0-point coordinates through the same part of the blurred comparison image information and the front and back image information, and obtain the coordinates of the boundary of the blurred region relative to the two-dimensional coordinate map after overlapping the blurred comparison image information and the front and back image information.

[0042] S402. Mark the coordinates of the boundary of the blurred region on the blurred comparison image information.

[0043] S403. Display the blurred comparison image information with the marked coordinates.

[0044] S500. Mark the boxed difference region determined to be blank as a blank region, and mark the blank region on the closest front and back image information to obtain blank comparison image information.

[0045] S501. Calculate the coordinates of the image edge and the blank region in the blank comparison image information, and compare them with the corresponding coordinates of the closest front and back image information. Call the coordinates of the front and back image information corresponding to the blank region stored in the database, determine the 0-point coordinates through the same part of the blank comparison image information and the front and back image information, and obtain the coordinates of the boundary of the blank comparison image information and the blank region relative to the two-dimensional coordinate map after overlapping the blank comparison image information and the front and back image information.

[0046] S502. Compare the boundary coordinates of the blank comparison image information with the boundary coordinates of the front and back image information.

[0047] S503. If the coordinates of the edges of the two images are different, calculate the coordinate difference and display it. If the coordinates of the edges of the two images are the same, display the blank comparison image information and the coordinates of the blank area.

[0048] S600. If the same front and back image information is found, use this front and back image information as the closest front and back image information until the complete image is completely different from the found front and back image information.

[0049] S700. After confirming the closest front and back image information, when receiving the complete image, compare the complete image with the determined front and back image information to determine whether the received complete image is the same as the determined front and back image information.

[0050] S701. If they are different, find the difference area between the received complete image and the determined front and back image information. If the received complete image is completely different from the determined front and back image information, issue an alarm and search for the same front and back image information in the database again. Confirming the closest front and back image information can confirm the product corresponding to this roll of film, making the subsequent judgment process faster and consuming less computing power. If there are other patterns, it means the product has changed, and the system can automatically switch the image information for judgment without manual operation by the user.

[0051] The implementation principle of a visual detection method applied to a bag-making machine in an embodiment of this application is as follows: During the working process of the bag-making machine, this system can automatically capture images of the front and back sides of the film, and can automatically determine the pattern and corresponding number of the current product through the database. If a printing error is found, it will automatically judge the closest product, and judge whether the printing is blurred or blank according to the pattern of the product itself. If the printing is blurred, mark the blurred position. If it is blank, first judge whether the entire pattern is printed out of position. If there is a printing misalignment, display the misalignment amount, otherwise display the marked blank position to facilitate the user to judge the printing fault. The user can always know through the system whether there is a printing error in the bag-making machine, and whether the printing error is pattern blurring, partial area blanking or printing layout misalignment.

[0052] Embodiment 2. An embodiment of this application discloses a visual detection system applied to a bag-making machine, as Figure 1 shown, including a data storage module 1, an image acquisition module 2, an image detection module 3, an error detection module 4, a blur judgment module 5, a blank judgment module 6, a fault display module 7, and a fault emergency stop module 8.

[0053] As Figure 1As shown in the figure, the data storage module 1 stores the front and back image information and the two-dimensional coordinate map according to the number. Each piece of front and back image information is established with coordinates based on the two-dimensional coordinate map. The image acquisition module 2 includes two image capture devices arranged on the bag-making machine. The image capture device can be an industrial camera. The two image capture devices respectively capture the complete images of the two layers of film, upload the front image information and the back image information, and transmit the front image information and the back image information to the image detection module 3.

[0054] As Figure 1 shown in the figure, when the image detection module 3 receives the complete image for the first time, it calls the front and back image information stored in the data storage module 1, and searches for the same front and back image information according to the front image information and the back image information. If the same front and back image information cannot be found, an alarm is issued and error information is transmitted to the error detection module 4. If the same front and back image information is found, the front and back image information is used as the closest front and back image information until the complete image is completely different from the found front and back image information. The alarm can include a sound alarm, a flash alarm, an information alarm, etc., and can be issued through an audible and visual alarm or the mobile terminal of the staff. After the image detection module 3 confirms the closest front and back image information, when receiving the complete image, it compares the complete image with the determined front and back image information to judge whether the received complete image is the same as the determined front and back image information. If they are not the same, it searches for the difference area between the received complete image and the determined front and back image information. If the received complete image is completely different from the determined front and back image information, an alarm is issued and the same front and back image information is searched again in the data storage module 1. Confirming the closest front and back image information can not only confirm the product corresponding to this roll of film, making the subsequent judgment process faster and consuming less computing power. If there are other patterns, it means that the product has changed, and the system can automatically switch the image information for judgment without manual operation by the user.

[0055] As Figure 1 shown in the figure, the fault emergency stop module 8 receives the alarm from the image detection module 3. The fault emergency stop module 8 presets the number of emergency stop triggers. If the number of consecutive alarms reaches the number of emergency stop triggers, it controls the bag-making machine to stop. If it receives an inspection instruction, it controls the bag-making machine to move the film until the same front and back image information is found, and then controls the bag-making machine to work normally. After consecutive multiple alarms are triggered, it can be confirmed that the entire roll of film has problems. At this time, the system automatically controls the bag-making machine to stop to avoid greater losses caused by continued processing. The staff can also manually input a detection instruction to make the bag-making machine only move the film. After it is determined that the pattern is intact, the processing ability of the bag-making machine will be automatically started to continue normal production.

[0056] As Figure 1As shown, the error detection module 4 presets a fuzzy judgment length. The error detection module 4 calls the information stored in the data storage module 1, searches for the closest front and back image information according to the error information, compares and locates the difference area between the received complete image and the closest front and back image information and frames the difference area, determines whether the image in the framed difference area is blurred, and at the same time determines whether the image in the framed difference area is blank. The error detection module 4 identifies the pixel cells of the image in the difference area, divides the pixel boundaries of the image in the difference area, determines the parts of the image pixel boundaries that are line segments or arcs as regular boundaries, and determines the remaining image pixel boundaries as irregular boundaries, and determines whether the length of the irregular boundaries exceeds the fuzzy judgment length. If the length of the irregular boundaries exceeds the fuzzy judgment length, it is determined that the image in the difference area is blurred. Mark the framed difference area determined to be blurred as a blurred area, mark the blurred area on the closest front and back image information to obtain blurred comparison image information, and transmit the blurred comparison image information to the fuzzy judgment module 5; mark the framed difference area determined to be blank as a blank area, mark the blank area on the closest front and back image information to obtain blank comparison image information, and transmit the blank comparison image information to the blank judgment module 6.

[0057] As Figure 1 shown, the fuzzy judgment module 5 calculates the coordinates of the blurred area and transmits the blurred comparison image information with coordinates to the fault display module 7. The fuzzy judgment module 5 calls the coordinates of the front and back image information corresponding to the blurred area stored in the data storage module 1, determines the 0-point coordinates through the same part of the blurred comparison image information and the front and back image information, obtains the coordinates of the boundary of the blurred area relative to the two-dimensional coordinate map after overlapping the blurred comparison image information and the front and back image information, marks the coordinates of the boundary of the blurred area on the blurred comparison image information, and transmits the marked blurred comparison image information to the fault display module 7.

[0058] As Figure 1 shown, the blank judgment module 6 calculates the coordinates of the image edge and the blank area in the blank comparison image information and compares them with the corresponding coordinates of the closest front and back image information. The blank judgment module 6 calls the coordinates of the front and back image information corresponding to the blank area stored in the data storage module 1, determines the 0-point coordinates through the same part of the blank comparison image information and the front and back image information, obtains the coordinates of the boundary of the blank comparison image information and the blank area relative to the two-dimensional coordinate map after overlapping the blank comparison image information and the front and back image information, and compares the boundary coordinates of the blank comparison image information with the boundary coordinates of the front and back image information. If the coordinates of the image edges of the two are different, calculate the coordinate difference and transmit the coordinate difference and the blank comparison image information to the fault display module 7. If the coordinates of the image edges of the two are the same, transmit the blank comparison image information with coordinates to the fault display module 7.

[0059] As Figure 1 shown, after receiving the image information, the fault display module 7 performs display. The bag making machine can be connected to a display terminal, which can be a desktop computer, a tablet computer, etc., and has functions of image display and information input.

[0060] The embodiments of this specific implementation manner are all preferred embodiments of the present invention, and do not limit the protection scope of the present invention accordingly. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A visual inspection method applied to a bag making machine, characterized in that: The following steps are also included: Establish a database to store the front and back image information and the two-dimensional coordinate map according to the number, and establish the coordinates of each front and back image information based on the two-dimensional coordinate map; Take complete images of two layers of film to obtain front and back image information; When receiving the complete image for the first time, searching for identical front and back image information in the database according to the front image information and the back image information; If the same front and back image information cannot be found, an alarm is issued and the closest front and back image information is searched in the database, and the difference area between the received complete image and the closest front and back image information is compared and selected; Determine whether the image in the selected difference area is blurred, and determine whether the image in the selected difference area is blank; Marking the framed difference area judged to be blurred as a blurred area, marking the blurred area on the closest front and back image information to obtain blurred contrast image information, calculating the coordinates of the blurred area, displaying the blurred contrast image information and displaying the coordinates; Mark the selected difference area judged as blank as a blank area, mark the blank area on the closest front and back image information to obtain blank contrast image information, calculate the coordinates of the image edge and the blank area in the blank contrast image information, and compare them with the corresponding coordinates of the closest front and back image information. If the coordinates of the edges of the two images are different, calculate the coordinate difference and display it. If the coordinates of the edges of the two images are the same, display the blank contrast image information and display the coordinates of the blank area. If the same front and back image information is found, the front and back image information is used as the closest front and back image information until the complete image is completely different from the found front and back image information.

2. A visual inspection method for a bag making machine according to claim 1, characterized in that: The following steps are also included: Set the number of emergency stop triggers; If the number of consecutive alarms reaches the emergency stop trigger number, the bag making machine is controlled to stop. If an inspection instruction is received, the bag making machine is controlled to move the film until the same front and back image information is found, and then the bag making machine is controlled to work normally.

3. The visual inspection method for bag making machine according to claim 1, characterized in that: The step of "determining whether the image in the selected distinguishing area is blurred" also includes: Set the fuzzy judgment length; Identify pixel cells of the image within the distinguished area and divide the image pixel boundaries within the distinguished area; The portion of the image pixel boundary that is a line segment or an arc is judged as a regular boundary, and the remaining image pixel boundaries are judged as irregular boundaries; Determine whether the length of the irregular boundary exceeds the fuzzy judgment length; If the length of the irregular boundary exceeds the blur judgment length, the image in the distinguished area is judged to be blurred.

4. The visual inspection method for bag making machine according to claim 1, characterized in that: The step of "calculating the coordinates of the fuzzy area, displaying the fuzzy contrast image information and displaying the coordinates" also includes: The coordinates of the front and back image information corresponding to the fuzzy area stored in the database are called, the 0-point coordinates are determined by the same part of the fuzzy contrast image information and the front and back image information, and the coordinates of the boundary of the fuzzy area on the two-dimensional coordinate map are obtained after the fuzzy contrast image information is overlapped with the front and back image information; Marking the coordinates of the fuzzy area boundary on the fuzzy contrast image information; Display the fuzzy contrast image information after marking the coordinates; The step of "calculating the coordinates of the image edge and the blank area in the blank contrast image information, and comparing them with the corresponding coordinates of the closest front and back image information" also includes; The coordinates of the front and back image information corresponding to the blank area stored in the database are called, and the 0-point coordinates are determined by the same part of the blank contrast image information and the front and back image information, and the coordinates of the boundary of the blank contrast image information and the blank area on the relative two-dimensional coordinate map are obtained after the blank contrast image information is overlapped with the front and back image information; The boundary coordinates of the blank contrast image information are compared with the boundary coordinates of the front and back image information.

5. The visual inspection method for bag making machine according to claim 1, characterized in that: The following steps are also included: After confirming the closest front and back image information, when receiving the complete image, compare the complete image with the determined front and back image information to determine whether the received complete image is the same as the determined front and back image information; If they are not the same, the difference area between the received complete image and the determined front and back image information is searched. If the received complete image is completely different from the determined front and back image information, an alarm is issued and the same front and back image information is searched again in the database.

6. A visual inspection system for a bag making machine, characterized in that: It comprises a data storage module (1), an image acquisition module (2), an image detection module (3), an error detection module (4), a fuzzy judgment module (5), a blank judgment module (6) and a fault display module (7); The data storage module (1) stores the front and back image information and the two-dimensional coordinate map according to the number, and each front and back image information has the coordinates established based on the two-dimensional coordinate map; The image acquisition module (2) comprises two image capturing devices arranged on the bag making machine, the two image capturing devices respectively capturing complete images of the two layers of film, uploading front image information and back image information, and transmitting the front image information and back image information to the image detection module (3); When the image detection module (3) receives the complete image for the first time, it calls the front and back image information stored in the data storage module (1), searches for identical front and back image information based on the front image information and the back image information, and if the identical front and back image information cannot be found, it issues an alarm and transmits error information to the error detection module (4); if the identical front and back image information is found, it uses the front and back image information as the closest front and back image information, until the complete image is completely different from the found front and back image information; The error detection module (4) calls the information stored in the data storage module (1), searches for the closest front and back image information according to the error information, compares and searches for the difference area between the received complete image and the closest front and back image information and selects the difference area, determines whether the image in the selected difference area is blurred, and determines whether the image in the selected difference area is blank, marks the selected difference area that is determined to be blurred as a blurred area, marks the blurred area on the closest front and back image information to obtain blurred contrast image information, and transmits the blurred contrast image information to the blurred judgment module (5); marks the selected difference area that is determined to be blank as a blank area, marks the blank area on the closest front and back image information to obtain blank contrast image information, and transmits the blank contrast image information to the blank judgment module (6); The fuzzy judgment module (5) calculates the coordinates of the fuzzy area and transmits the fuzzy contrast image information with the coordinates to the fault display module (7); The blank judgment module (6) calculates the coordinates of the image edge and the blank area in the blank contrast image information, and compares them with the corresponding coordinates of the closest front and back image information. If the coordinates of the image edges of the two images are different, the coordinate difference is calculated, and the coordinate difference and the blank contrast image information are transmitted to the fault display module (7). If the coordinates of the image edges of the two images are the same, the blank contrast image information with the coordinates is transmitted to the fault display module (7); The fault display module (7) receives the image information and then displays it.

7. The visual inspection system for bag making machine according to claim 6, characterized in that: It also includes a fault emergency stop module (8), which receives the alarm from the image detection module (3). The fault emergency stop module (8) presets the emergency stop triggering number. If the number of consecutive alarms reaches the emergency stop triggering number, the bag making machine is controlled to stop. If an inspection instruction is received, the bag making machine is controlled to move the film until the same front and back image information is found, and then the bag making machine is controlled to work normally.

8. The visual inspection system for bag making machine according to claim 6, characterized in that: The error detection module (4) presets a blur judgment length, identifies pixel cells of the image in the distinguished area, divides the image pixel boundaries in the distinguished area, judges the portion of the image pixel boundaries that is a line segment or an arc as a regular boundary, judges the remaining image pixel boundaries as an irregular boundary, judges whether the length of the irregular boundary exceeds the blur judgment length, and judges the image in the distinguished area as blurry if the length of the irregular boundary exceeds the blur judgment length.

9. The visual inspection system for bag making machine according to claim 6, characterized in that: The fuzzy judgment module (5) calls the coordinates of the front and back image information corresponding to the fuzzy area stored in the data storage module (1), determines the zero point coordinates through the same portion of the fuzzy contrast image information as the front and back image information, obtains the coordinates of the boundary of the fuzzy area on the two-dimensional coordinate diagram after the fuzzy contrast image information is overlapped with the front and back image information, marks the coordinates of the boundary of the fuzzy area on the fuzzy contrast image information, and transmits the marked fuzzy contrast image information to the fault display module (7); The blank judgment module (6) calls the coordinates of the front and back image information corresponding to the blank area stored in the data storage module (1), determines the zero point coordinates through the same portion of the blank contrast image information as the front and back image information, overlaps the blank contrast image information with the front and back image information, obtains the coordinates of the boundary of the blank contrast image information and the blank area on the two-dimensional coordinate diagram, and compares the boundary coordinates of the blank contrast image information with the boundary coordinates of the front and back image information.

10. The visual inspection system for bag making machine according to claim 6, characterized in that: After the image detection module (3) confirms the closest front and back image information, when receiving the complete image, it compares the complete image with the determined front and back image information to determine whether the received complete image is the same as the determined front and back image information. If they are not the same, it searches for the difference area between the received complete image and the determined front and back image information. If the received complete image is completely different from the determined front and back image information, it issues an alarm and searches the data storage module (1) again for the same front and back image information.

Images