An interactive user feedback system applied to promote improvement of inspection accuracy of an automated visual inspection system

Abstract

To provide an interactive user feedback system applied to enhancement of the inspection accuracy of an automated visual inspection system.SOLUTION: When working normally, an automated visual inspection (AVI) system acquires an article image from a continuous article transferred by a transfer device, and then determines whether or not there is at least one defect existing in the article image. Subsequently, the interactive user feedback system 1 enables a display screen 11D of an electronic device 11 to develop an inspection report section including M×N sub sections. A quality inspector clicks one sub section to cause the display screen to display one sub image section of the article image corresponding to the sub section. The AVI system inspects whether or not a defect determination result created for the sub image section is correct. If not, the quality inspector is able to revise the defect determination result and realize the user feedback.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to the technical field of machine vision, and more particularly to a user feedback system applied to promote the improvement of inspection accuracy of an automatic visual inspection system. [Background technology]

[0002] It is already known that after a textile (e.g., a fabric) is produced, it is necessary to receive a defect inspection in advance before shipping and packaging. Currently, the defect inspection of a fabric is usually completed by a quality inspector operating an inspection machine. The main structure of a conventional fabric inspection machine includes a material release mechanism, a fabric inspection platform, and a material receiving mechanism. When inspecting the fabric, a quality inspector stands in front of the operating platform and controls the fabric release speed of the material release mechanism and the fabric winding speed of the material receiving mechanism by operating a controller. In the process of controlling the movement of the fabric, the quality inspector visually inspects the fabric section stretched on the fabric inspection platform. If the quality inspector finds that the fabric section contains a defect, he operates the controller to stop the movement of the fabric. Then, the quality inspector judges whether the defect is repairable, and if it is repairable, he immediately repairs the defect, and if it is not repairable, he marks the position of the defect.

[0003] Traditional fabric inspection machines have the following major defects in practical use: quality inspectors use their naked eyes to visually detect whether there are defects in the fabric sections spread out on the fabric inspection platform during the fabric moving process, and this method will inevitably lead to the existence of defects in the fabric that are not detected by human eyes. In view of the above, a defect inspection system with motor control function (also referred to as an automatic visual inspection system) disclosed in Patent Document 1 is used for alignment in the above-mentioned fabric inspection machine, and includes a linear light source, at least one imager, and a processing and control module. During normal operation, the inspection light provided by the linear light source is projected within the imaging range of the imager, and the processing and control module controls the motors of the material release mechanism and the material receiving mechanism to operate, thereby moving the fabric so as to be conveyed. When the fabric moves to position one of the inspection sections within the imaging range, the imager can capture the inspection section to obtain a fabric image. Then, the processing and control module performs a defect inspection on the fabric image, and checks whether at least one defect exists in the inspection section. If the defect is found, the process and control module then controls the motors of the material release mechanism and the material receiving mechanism to operate, so as to move the fabric to be conveyed and place the inspection area on the inspection platform. Finally, the quality inspector determines whether the defect is repairable or not, and if so, repairs the defect immediately, or marks the defect location if not.

[0004] More specifically, the processing and control module has a pre-trained defect inspection model and can utilize the defect inspection model to perform the defect inspection on the fabric image. Any AI engineer skilled in image recognition technology will know that the defect inspection model is a model generated after performing machine learning training of fabric defect inspection (or fabric defect identification) on an image classifier using a plurality of training samples, among which the plurality of training samples include a plurality of normal fabric images (i.e., golden samples) and a plurality of fabric images containing at least one defect, and the fabric images containing these defects can be pre-classified into a plurality of reference defect feature images having creases, tears, color stains, houndstooth patterns, wavy patterns, warp defects, weft defects, etc. According to this setup, the defect inspection model first performs feature extraction processing on the fabric images to obtain fabric feature images. The defect inspection model then determines whether the inspected area of ​​the fabric contains a defect defined by an inspection specification by performing a feature comparison against the fabric feature image and at least one reference defect feature image stored in the database.

[0005] However, according to the use experience of quality inspectors, the fabric defects detected and classified by the defect inspection model are not always completely accurate. Figure 1 is a Venn diagram of defect verification results. As shown in Figure 1, area A is used to display the statistical data range (abbreviated as "missed judgment range") in which the quality inspector judges the fabric to have a defect, but the automated visual inspection system judges the fabric to have no defect. Meanwhile, area B is used to display the statistical data range (abbreviated as "candidate judgment range") in which the quality inspector judges the fabric to have no defect, but the automated visual inspection system judges the fabric to have a defect. In addition, there is an area C between area A and area B, and the area C is a common set area between area A and area B. It should be understood that the area C is used to display the statistical data range (abbreviated as the judgment candidate range) that is judged by the quality inspector as having a defect in the fabric, but is also judged by the automatic visual inspection system as having a defect in the fabric. Therefore, as can be seen from Fig. 1, although the current automatic visual inspection system can automatically detect defects occurring in the fabric, the detected defects are not always completely accurate. The main reason is that there are many classification items of fabric defects, and therefore, when the number of fabric images containing defects included in the database is insufficient, there are frequent cases where defect A is misjudged as defect B due to insufficient training on the defect inspection model. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Taiwan Patent No. I767828(B) Specification Summary of the Invention [Problem to be solved by the invention]

[0007] In view of the above, a user feedback system has been developed in consideration of the above, which is integrated in an electronic device (e.g., a tablet computer), and the processing and control module allows the fabric image and related defect detection (judgment) results to be displayed on the screen of the tablet computer for quality inspectors to check and confirm. For this reason, the inventor of the present invention has researched and devised as much as possible, and finally researched, developed and completed a two-way user feedback system applied to improving and promoting the inspection accuracy of the automatic visual inspection system according to the present invention.

[0008] The main object of the present invention is to provide an interactive user feedback system, which is realized by using an electronic device, for promoting and improving the inspection accuracy of an automatic visual inspection (AVI) system. During normal operation, the AVI system can take an article image from a continuous material conveyed by a conveying device, and determine whether the article image contains at least one defect. Then, the interactive user feedback system of the present invention causes an inspection report section including M×N subsections to appear on the display screen of the electronic device. Thus, a quality inspector can click on a subsection to display one sub-image section of the article image corresponding to the subsection on the display screen, and then the AVI system checks whether the defect judgment result made for this sub-image section is accurate. If the result is no, the quality inspector can make corrections to the defect judgment result, and the article image after receiving the corrections can be stored in a training sample database, thereby realizing user feedback.

[0009] For example, the interactive user feedback system of the present invention may be an application program installed in a tablet computer (i.e., the electronic device described above), and a quality inspector in a manufacturing plant can operate the tablet computer to modify the defect inspection result of a continuous material in any one of the AVI systems in the plant. Furthermore, the output manner of the article image that has received the defect classification modification by the quality inspector may be stored in a system database and / or uploaded to a remote electronic device. Thus, the remote electronic device can generate at least one update module using a plurality of normal article images and a plurality of defective article images including at least one article image that has already received the defect classification modification, and then uses the update module to perform a model update operation on a first article defect identification model installed in the main control electronic device, or uses the update module to install a second article defect identification model in the main control electronic device. [Means for solving the problem]

[0010] In order to achieve the above object, the present invention provides an interactive user feedback system in one embodiment, which is used for promoting and improving the inspection accuracy of an automatic visual inspection system, wherein the automatic visual inspection system is matched with a continuous material conveying device having at least one roll-to-roll equipment, and includes a plurality of imaging devices and a main control electronic device, wherein the continuous material conveying device conveys a continuous flexible material (continuous flexible material), the main control electronic device performs the defect inspection operation on the article image, and then identifies and classifies K defects present in the article image, where K is a positive integer; the interactive user feedback system comprises an electronic device electrically connected to the main control electronic device, the electronic device including a first processor, a first memory, and a first display screen, wherein the first processor stores a first application program in the first memory, and the first processor executes the first application program through access to the first memory to enable the following functions: The system receives a long side range and a short side range of the actual inspection section of the continuous material, generates an inspection report section corresponding to the item image based on the long side range and the short side range, and displays the inspection report section divided into M x N subsections on the first display screen, where M and N are positive integers, and after one of the M x N subsections displayed on the first display screen is clicked, displays on the first display screen one sub-image section of the item image corresponding to the subsection, a defect inspection report for the sub-image section, and a defect type selection menu, and after the defect type selection menu is clicked, a defect classification correction is made to the defect inspection report of the sub-image section, and the corrected defect inspection report and item image are stored.

[0011] In one embodiment, the electronic device is any one selected from the group consisting of a smartphone, a tablet computer, a desktop computer, an all-in-one computer, and a notebook computer.

[0012] In one embodiment, the continuous material is any one selected from the group consisting of a metal foil, a metal sheet, a flexible touch panel, a flexible circuit board, and a flexible solar cell.

[0013] In one executable embodiment, the first processor, in accordance with executing the first application program, further enables the following function: upload at least one item image to a remote electronic device, so that the remote electronic device can generate at least one update module using a plurality of normal item images and a plurality of the item images, and then use the update module to perform a model update operation on a first item defect identification model installed in the main control electronic device, or use the update module to install a second item defect identification model in the main control electronic device.

[0014] In one embodiment, the automatic visual inspection system further includes a length measuring device electrically connected to the main control electronic device, and the length measuring device continuously records the article movement distance during the process of transporting the continuous material, so that the main control electronic device can calculate the long side range and the short side range obtained based on the article movement distance and article basic information when the actual inspection area of ​​the continuous material enters the shooting area.

[0015] In one embodiment, each of the sub-regions has a region coordinate, and the region coordinate includes a long side coordinate value and a short side coordinate value.

[0016] In one embodiment, the roll-to-roll equipment includes a material releasing mechanism, a material receiving mechanism, and a verification platform, and the main control electronic device includes a main processor and a main memory, in which a second application program is stored in the main memory, and the main processor executes the second application program through access to the main memory to enable the following functions and control the material releasing mechanism and the material receiving mechanism to operate to transport the continuous material, and during the process of transporting the continuous material, the plurality of imaging devices capture images of the front of the actual inspection section of the continuous material in the shooting area. The apparatus controls to photograph the object image, performs a feature comparison operation on the object image and a plurality of reference defect feature images to determine whether the object image contains at least one defect, and if it is determined that the object image contains at least one defect, controls the material releasing mechanism and the material receiving mechanism to transport the continuous material to position the actual inspection area on the verification platform, then stops the operation of the material releasing mechanism and the material receiving mechanism, and controls to operate the material releasing mechanism and the material receiving mechanism to transport the continuous material after completing the defect classification correction.

[0017] In one embodiment, the roll-to-roll equipment further includes a first rotation rate sensor connected to a first motor of the material discharge mechanism and used to sense a first rotation rate of the first motor, and a second rotation rate sensor connected to a second motor of the material receiving mechanism and used to sense a second rotation rate of the second motor.

[0018] In one embodiment, the main control electronic device further includes a man-machine interface, and can operate the man-machine interface to send a restart command to the main processor, and the main processor drives the first motor and the second motor after completing the defect classification correction, thereby operating the material releasing mechanism and the material receiving mechanism to continuously transport the continuous material.

[0019] In one embodiment, each of the imagers has an imaging range, the plurality of imagers has a total imaging range, and the width of the total imaging range is greater than or equal to the width of the web. [Brief description of the drawings]

[0020] [Figure 1] This is a Venn diagram of the defect verification results. [Diagram 2] 1 is a block diagram of an automated vision inspection system including an interactive user feedback system in accordance with the present invention; [Figure 3A] FIG. 1 is a first perspective view of a continuous material conveying apparatus including an automated visual inspection system. [Figure 3B] FIG. 2 is a second perspective view of the continuous material transport device. [Figure 4] FIG. 13 illustrates an example of an examination report section. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] In order to more clearly describe the interactive user feedback system applied to promote the improvement of inspection accuracy of the automated visual inspection system proposed by the present invention, the preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] FIG. 2 is a block diagram of an automated visual inspection system including an interactive user feedback system according to the present invention. As shown in FIG. 2, the interactive user feedback system 1 according to the present invention is realized by using an electronic device 11, and is used to promote and improve the inspection accuracy of an automated visual inspection system 3. Furthermore, FIG. 3A and FIG. 3B are a first perspective view and a second perspective view of a continuous material conveying device including an automated visual inspection system. As shown in FIG. 2, FIG. 3A and FIG. 3B, the automated visual inspection system 3 is aligned with a continuous material conveying device 2 including at least one roll-to-roll equipment, and includes a light source 31, a plurality of cameras 32, and a main control electronic device 33. During normal operation, the continuous material conveying device 2 conveys a continuous flexible material, and when the actual inspection area of ​​the continuous material enters the shooting area, the multiple cameras 32 take article images of the continuous material, and then the main control electronic device 33 performs a defect inspection operation on the article image, and then identifies and classifies K defects present in the article image, where K is a positive integer.

[0023] It should be emphasized that the continuous material conveying device 2 illustrated in Figures 3A and 3B is a fabric inspection machine, so the continuous material is a fabric. However, in a possible embodiment, the continuous material conveying device 2 does not need to be limited to a fabric inspection machine, and it may be other types of continuous material conveying devices 2 used to convey metal foil, metal thin sheet, flexible touch panel, flexible printed circuit board, and flexible solar cell.

[0024] 3A and 3B, the main mechanisms of the continuous material conveying device 2 include a material releasing mechanism 21, a material receiving mechanism 23, and a verification platform 22. Meanwhile, the main control electronic device 33 includes a main processor 33P and a main memory 33M, in which an automatic vision inspection application program is stored in the main memory 33M, so that the main processor 33P can execute the automatic vision inspection application program through access to the main memory 33M. According to this installation, when a defect inspection is performed on a roll of continuous material using the automatic visual inspection system 3, the main processor 33P executes an apparatus control program included in the automatic visual inspection application program to control the operation of the material releasing mechanism 21 and the material receiving mechanism 23. Meanwhile, an article defect identification program is also installed in the main control electronic device 33, which is used to perform a feature comparison operation on the article image of the continuous material captured by the multiple cameras 32 and multiple reference defect feature images, and determines whether the article image contains at least one defect.

[0025] More specifically, during the process of transporting the continuous material, the multiple imagers 32, under the control of the main processor 33P, photograph the continuous material to obtain an article image when the substance inspection area of ​​the continuous material enters the photographing area. Since the continuous material includes a plurality of continuous substance inspection areas, it should be understood that after the material receiving mechanism 23 completely receives one roll of the continuous material supplied and discharged from the material discharging mechanism 21, the multiple imagers 32 can take a plurality of article images corresponding to the plurality of substance inspection areas. Meanwhile, Figures 3A and 3B also show that a support mechanism 35 is installed in the photographing area. In practical application, the support mechanism 35 is used to support the substance inspection area of ​​the continuous material, and the substance inspection area is flatly spread out in the photographing area.

[0026] As shown in Figures 3A and 3B, each of the imagers 32 has an image range, the multiple imagers 32 have a total image range, and the width of the total image range is greater than or equal to the width of the object. According to this configuration, one image of an object captured using the multiple imagers 32 includes one complete object inspection area. After obtaining the article image, the main control electronic device 33 performs a feature comparison operation on the article image and a plurality of reference defect feature images to determine whether the article image contains at least one defect. If it is determined that the article image contains at least one defect, the main control electronic device 33 controls the material releasing mechanism 21 and the material receiving mechanism 23 to transport the continuous material to position the actual inspection area on the verification platform 22, and then stops the operation of the material releasing mechanism 21 and the material receiving mechanism 23.

[0027] According to the design of the present invention, as shown in FIG. 2, FIG. 3A and FIG. 3B, the interactive user feedback system 1 includes an electronic device 11 and a user feedback program installed in the electronic device 11. Specifically, the electronic device 11 may be a smartphone, a tablet computer, a desktop computer, an all-in-one computer or a notebook computer. The electronic device 11 is electrically connected to the main control electronic device 33 and includes a first processor 11P, a first memory 11M and a first display screen 11D. In one embodiment, by storing the user feedback program in the first memory 11M, the first processor 11P causes a user interface (UI) to be presented on the first display screen 11D of the electronic device 11 according to executing the user feedback program through access to the first memory 11M.

[0028] After the automatic visual inspection system 3 completes the defect inspection work of one article image, the automatic visual inspection system 3 generates a defect inspection report. After the electronic device 11 receives the article image and the defect inspection report from the main control electronic device 33, the first processor 11P generates an inspection report section corresponding to the article image according to the long side range and the short side range of the actual inspection section of the continuous material. FIG. 4 is a diagram showing an example of the inspection report section. As shown in FIG. 4, the inspection report section is displayed in the user interface of the first display screen 11D, and the inspection report section includes M×N sub-sections, where M and N are positive integers.

[0029] To further explain, the continuous material conveying device 2 further includes a first rotation rate sensor 211 and a second rotation rate sensor 231, among which the first rotation rate sensor 211 is connected to the first motor 210 of the material discharge mechanism 21 and is used to detect the first rotation rate of the first motor 210. Meanwhile, the second rotation rate sensor 231 is connected to the second motor 230 of the material receiving mechanism 23 and is used to detect the second rotation rate of the second motor 230. Therefore, the main processor 33P can control the first motor 210 and the second motor 230 to operate at a specified rotation rate through the execution of a device control program included in the automatic visual inspection application program, and in this manner, the moving speed of the continuous material is controlled. Meanwhile, the automatic visual inspection system 3 also includes a length measuring device 30 electrically connected to the main control electronic device 33, and the length measuring device 30 is used to continuously record the moving distance of the article during the process of conveying the continuous material.

[0030] 3A, 3B and 4, when the object area of ​​the continuous material is in the shooting area, the main control electronic device 33 can calculate the long side range and short side range of the object area of ​​the continuous material in the shooting area based on the moving distance of the article and the basic information of the article, so that the first processor 11P of the first electronic device 11 can then generate an inspection report area including M×N sub-areas based on the long side range and the short side range. As shown in FIG. 4, each of the sub-areas has an area coordinate (e.g., A5, B4, C3, D2, E1, F0), and the area coordinate consists of a long side coordinate value (i.e., A, B, C, ..., I, J) and a short side coordinate value (i.e., 5, 4, 3, 2, 1, 1).

[0031] According to the design of the present invention, after the inspection report section is displayed on the first display screen 11D, the quality inspector can select and click one of the M×N subsections. Furthermore, after one of the M×N subsections displayed on the first display screen 11D is clicked, the first display screen 11D displays one sub-image section of the item image corresponding to the subsection. In other words, the M×N subsections are M×N sub-image sections corresponding to the item images, respectively. Moreover, when one sub-image section is displayed on the first display screen 11D, the first display screen 11D simultaneously displays a defect inspection report of the sub-image section, in which the defect inspection report includes the quantity and type of defects contained in the sub-image section. In addition, the first display screen 11D also displays a defect type selection menu (or a list table).

[0032] At this time, the quality inspector can check with the naked eye whether the defect feature is really included in the sub-image section, and if there is no defect feature, the quality inspector can click other selection items in the defect type selection menu to perform a defect classification correction operation on the defect inspection report of the sub-image section. On the other hand, if the quality inspector specifies the defect type of the sub-image section from the defect inspection report, the quality inspector can click and select the correct defect type in the defect type selection menu to perform a defect classification correction operation on the defect inspection report of the sub-image section, different from what can actually be confirmed with the naked eye. Finally, after the quality inspector checks the sub-image sections and their defect inspection reports corresponding to each sub-section one by one, he operates the electronic device 11 to store the corrected defect inspection report and the item image, and then operates the man-machine interface of the main control electronic device 33 to send a restart command to the main processor 33P, and the main processor 33P drives the first motor 210 and the second motor 230 to operate the material release mechanism 21 and the material receiving mechanism 23 to continuously transport the continuous material.

[0033] Further, the electronic device 11 can also be operated to upload a plurality of article images and a plurality of the defect inspection reports corresponding to the actual inspection sections of the plurality of portions of the continuous material to the remote electronic device 4. In this way, the remote electronic device 4 can generate at least one update module using a plurality of normal article images and a plurality of the article images. Then, the electronic device 11 can be operated to use the update module to perform a model update operation on the article defect identification model installed in the main control electronic device 33, or use the update module to install another set of article defect identification models in the main control electronic device 33.

[0034] In summary, the present invention provides an interactive user feedback system that is realized using electronic devices and is used to promote and improve the inspection accuracy of an automated visual inspection (AVI) system. During normal operation, the AVI system can take an article image from a continuous material conveyed by a conveying device, and determine whether the article image contains at least one defect. Then, the interactive user feedback system of the present invention causes an inspection report section including M×N subsections to appear on the display screen of the electronic device. Thus, the quality inspector can click on a subsection to display a sub-image section of the article image corresponding to the subsection on the display screen, and then the AVI system checks whether the defect judgment result made for this sub-image section is correct. If the result is no, the quality inspector can make corrections to the defect judgment result, and the article image after receiving the corrections can be stored in the training sample database, thereby realizing user feedback.

[0035] The interactive user feedback system applied to promote the improvement of the inspection accuracy of the automated visual inspection system according to the present invention has been fully and clearly described above. However, it should be emphasized that the above detailed description is a concrete description of the feasible embodiments of the present invention, and the scope of the present invention is not limited to these embodiments, and the equivalent implementations or modifications thereof are still within the scope of the claims of the present application without departing from the technical spirit of the present invention. [Explanation of symbols]

[0036] 1. Two-way user feedback system 11 Electronic equipment 11P 1st processor 11M 1st memory 11D First display screen 2 Continuous material transport device 21 Material release mechanism 210 First Motor 211 First rotation rate sensor 22 Verification Platform 23 Material Receiving Mechanism 230 Second Motor 231 Second rotation rate sensor 3. Automated visual inspection system 30 Length measuring device 31 Light source 32 Camera 33 Main control electronics 33P main processor 33M main memory 35 Support mechanism 4 Remote Electronic Devices

Claims

1. 1. An interactive user feedback system for use in promoting and improving inspection accuracy of an automated visual inspection system, comprising: The automated visual inspection system is matched with a continuous material conveying device having at least one roll-to-roll facility, and includes a plurality of imagers and a main control electronic device, the continuous material is conveyed by the continuous material conveying device, so that when a real inspection area of ​​the continuous material enters a shooting area, the plurality of imagers can take an article image of the continuous material, and the main control electronic device performs a defect inspection operation on the article image, and then identifies and classifies K defects existing in the article image, where K is a positive integer; The interactive user feedback system comprises an electronic device electrically connected to the main control electronic device and including a first processor, a first memory, and a first display screen, storing a first application program in the first memory, the first processor enabling the following functions as it executes the first application program through access to the first memory: receiving from the master control electronic device a long side extent and a short side extent of the actual inspection section of the continuous material; generating an inspection report section corresponding to the article image based on the long side range and the short side range, and displaying the inspection report section divided into M×N subsections on the first display screen, where M and N are positive integers; After one of the M×N subsections displayed on the first display screen is clicked, a subimage section of one of the article images corresponding to the subsection, a defect inspection report for the subimage section, and a defect type selection menu are displayed on the first display screen; When the defect type selection menu is clicked, a defect classification correction is performed on the defect inspection report in the sub-image section, and then the corrected defect inspection report and the article image are stored; The first processor further enables the following functions as the first application program is executed: By uploading at least one of the article images to a remote electronic device, the remote electronic device can generate at least one update module using a plurality of normal article images and a plurality of the article images, and then use the update module to perform a model update operation on a first article defect identification model installed in the main control electronic device, or use the update module to install a second article defect identification model in the main control electronic device. An interactive user feedback system.

2. The interactive user feedback system of claim 1, wherein the electronic device is any one selected from the group consisting of a smartphone, a tablet computer, a desktop computer, an all-in-one computer, and a notebook computer.

3. The interactive user feedback system of claim 1 , wherein the continuous material is any one selected from the group consisting of a metal foil, a thin metal plate, a flexible touch panel, a flexible circuit board, and a flexible solar cell.

4. The interactive user feedback system of claim 1, characterized in that the automatic visual inspection system further includes a length measuring device electrically connected to the main control electronic device, and the length measuring device continuously records the article movement distance during the process of transporting the continuous material, so that the main control electronic device can calculate the long side range and the short side range obtained based on the article movement distance and article basic information when the actual inspection area of ​​the continuous material enters the shooting area.

5. The interactive user feedback system according to claim 1 , wherein each of the sub-sections has a section coordinate, and the section coordinate includes a long side coordinate value and a short side coordinate value.

6. A two-way user feedback system for use in promoting and improving inspection accuracy of an automated visual inspection system, comprising: The automated visual inspection system is matched with a continuous material conveying device having at least one roll-to-roll facility, and includes a plurality of imagers and a main control electronic device, the continuous material is conveyed by the continuous material conveying device, so that when a real inspection area of ​​the continuous material enters a shooting area, the plurality of imagers can take an article image of the continuous material, and the main control electronic device performs a defect inspection operation on the article image, and then identifies and classifies K defects existing in the article image, where K is a positive integer; The interactive user feedback system comprises an electronic device electrically connected to the main control electronic device and including a first processor, a first memory, and a first display screen, storing a first application program in the first memory, the first processor enabling the following functions as it executes the first application program through access to the first memory: receiving from the master control electronic device a long side extent and a short side extent of the actual inspection section of the continuous material; generating an inspection report section corresponding to the article image based on the long side range and the short side range, and displaying the inspection report section divided into M×N subsections on the first display screen, where M and N are positive integers; After one of the M×N subsections displayed on the first display screen is clicked, a subimage section of one of the article images corresponding to the subsection, a defect inspection report for the subimage section, and a defect type selection menu are displayed on the first display screen; When the defect type selection menu is clicked, a defect classification correction is performed on the defect inspection report in the sub-image section, and then the corrected defect inspection report and the article image are stored; the roll-to-roll equipment includes a material release mechanism, a material receiving mechanism, and a verification platform; The main control electronic device includes a main processor and a main memory, and stores a second application program in the main memory, and the main processor executes the second application program through access to the main memory to enable the following functions: controlling the material discharge mechanism and the material receiving mechanism to operate to transport the continuous material; During the process of transporting the continuous material, the plurality of image capturing devices are controlled to capture the object images of the actual inspection area of ​​the continuous material within the image capturing area; performing a feature comparison operation on the article image and a plurality of reference defect feature images to determine whether the article image contains at least one defect; If it is determined that the article image contains at least one defect, controlling the material releasing mechanism and the material receiving mechanism to transport the continuous material to position the actual inspection section on the verification platform, and then stopping the operation of the material releasing mechanism and the material receiving mechanism; After completing the defect classification correction, the material discharge mechanism and the material receiving mechanism are controlled to operate so as to transport the continuous material. An interactive user feedback system.

7. 7. The interactive user feedback system of claim 6, wherein the continuous material transport device further includes a first rotation rate sensor connected to a first motor of the material discharge mechanism and used to sense a first rotation rate of the first motor, and a second rotation rate sensor connected to a second motor of the material receiving mechanism and used to sense a second rotation rate of the second motor.

8. The interactive user feedback system of claim 7, characterized in that the main control electronic device further includes a man-machine interface, and can send a re-operation command to the main processor by operating the man-machine interface, and the main processor drives the first motor and the second motor after completing the defect classification correction, thereby operating the material releasing mechanism and the material receiving mechanism to continuously transport the continuous material.

9. 9. The interactive user feedback system of claim 8, wherein each of the imaging devices has an imaging range, the imaging devices have a total imaging range, and the width of the total imaging range is greater than or equal to the width of the continuous material.

Images