Database construction method for tank visual inspection

Abstract

The present invention provides a method for building a database for visual inspection of can bodies, comprising: obtaining multiple groups of test empty cans; performing a single rejection test on a single test empty can in the multiple groups using a visual inspection device to obtain a single rejection test result; performing a continuous can rejection test on a plurality of continuously conveyed test empty cans in the multiple groups using the visual inspection device to obtain a continuous can rejection test result; and obtaining a database for visual inspection of can bodies based on the test results of the single rejection test and the test results of the continuous can rejection test. The technical solution of the present application effectively solves the problem of high randomness and easy missed detection in visual inspection in related technologies.

Description

Technical Field

[0001] The present invention relates to the technical field of empty can detection, and in particular to a method for building a database for visual inspection of can bodies. Background Art

[0002] Visual inspection is a crucial component of industrial production. Highly accurate machine vision inspection devices can automatically process large amounts of information. In intelligent manufacturing and automated production processes, machine vision systems are widely used in fields such as process monitoring, finished product inspection, and quality control. Visual inspection devices primarily use industrial cameras to replace the human eye in performing functions such as identification, measurement, and positioning. Typically, a visual inspection device consists of a camera, lens, and light source, replacing manual inspection for barcode characters, cracks, packaging integrity, and dents. Using visual inspection devices can effectively improve the speed and accuracy of production line inspections, significantly increasing output and quality, reducing labor costs, and preventing misjudgments caused by eye fatigue. In recent years, they have been widely used in industries such as food and pharmaceuticals. Milk powder processing companies use cans to store milk powder. Therefore, visual inspection is necessary to inspect these cans.

[0003] In related technologies, visual inspection lacks standardized simulated test samples, is highly random, and is prone to overlooking difficult-to-detect points inside the product. It cannot simulate the actual production state, and cannot achieve accurate effectiveness testing and equipment precision calibration. The entire debugging cycle is very long. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method for building a database for visual inspection of tank bodies, so as to solve the problem that visual inspection in related technologies is relatively random and prone to missed detection.

[0005] In order to achieve the above-mentioned objectives, the present invention provides a method for building a database for visual inspection of can bodies, comprising: obtaining multiple groups of test empty cans; using a visual inspection device to perform a single rejection test on a single test empty can in the multiple groups of test empty cans to obtain a single rejection test result; using a visual inspection device to perform a continuous can rejection test on multiple continuously conveyed test empty cans in the multiple groups of test empty cans to obtain a continuous can rejection test result; and obtaining a database for visual inspection of can bodies based on the test results of the single rejection test and the test results of the continuous can rejection test.

[0006] Furthermore, the steps of obtaining multiple groups of test empty cans include: placing foreign objects inside the can body to obtain a first group of test empty cans; squeezing and deforming the can body to obtain a second group of test empty cans; and shearing the mouth or bottom edge of the can body to obtain a third group of test empty cans.

[0007] Furthermore, the step of using a visual inspection device to perform a single rejection test on a single test empty can in multiple groups of test empty cans to obtain a single rejection test result includes: performing multiple tests on multiple test empty cans with different foreign matter positions in the first group of test empty cans, and performing multiple tests on each test empty can in the first group of test empty cans; performing multiple tests on multiple test empty cans with different shapes in the second group of test empty cans, and performing multiple tests on each test empty can in the second group of test empty cans; performing multiple tests on multiple test empty cans with different incision positions in the third group of test empty cans, and performing multiple tests on each test empty can in the third group of test empty cans.

[0008] Furthermore, the step of performing multiple tests on each test empty can in the first group of test empty cans includes: rotating the test empty can by a first preset angle or inverting the test empty can for each test; performing multiple tests on each test empty can in the second group of test empty cans: rotating the test empty can by a first preset angle or inverting the test empty can for each test; performing multiple tests on each test empty can in the third group of test empty cans: rotating the test empty can by a first preset angle or inverting the test empty can for each test.

[0009] Furthermore, the steps of using a visual inspection device to perform a continuous can rejection test on a plurality of continuously conveyed test empty cans in a plurality of groups of test empty cans to obtain a continuous can rejection test result include: continuously testing the first group of test empty cans by the visual inspection device; continuously testing the second group of test empty cans by the visual inspection device; and continuously testing the third group of test empty cans by the visual inspection device.

[0010] Furthermore, the step of using a visual inspection device to perform a continuous can rejection test on multiple continuously conveyed test empty cans in multiple groups of test empty cans to obtain a continuous can rejection test result includes: when continuous testing is performed using multiple groups of test empty cans and a visual inspection device, the latter test empty can of two adjacent test empty cans is rotated by a second preset angle relative to the previous test empty can.

[0011] Furthermore, the first preset angle is 30°, and the second preset angle is 60°.

[0012] Furthermore, when continuous testing is performed using multiple groups of test empty cans and visual inspection devices, the step of rotating the latter test empty can of two adjacent test empty cans relative to the former test empty can by a second preset angle includes: forming six test empty cans into a test set, and the rotation angle of the first test empty can in each test set is different.

[0013] Furthermore, the step of using a visual inspection device to perform a single rejection test on a single test empty can in multiple groups of test empty cans to obtain a single rejection database includes: recording the image data of all rejected test empty cans through the visual inspection device, and obtaining a single rejection test result based on the image data of all rejected test empty cans; the step of using a visual inspection device to perform a continuous can rejection test on multiple continuously conveyed test empty cans in multiple groups of test empty cans to obtain a continuous can rejection test result includes: recording the image data of all rejected test empty cans through the visual inspection device, and obtaining a continuous can rejection test result based on the image data of all rejected test empty cans.

[0014] Furthermore, the step of using a visual inspection device to perform a single rejection test on a single test empty can in multiple groups of test empty cans includes: recording the image data of all rejected test empty cans through the visual inspection device; when there are test empty cans that have not been rejected, testing the test empty cans that have not been rejected again, and controlling the visual inspection device to record the image data of the test empty cans that have not been rejected; recording the image data of all rejected test empty cans and the image data of the test empty cans that have not been rejected as the test result of the single rejection test; the step of using a visual inspection device to perform a continuous can rejection test on multiple test empty cans that are continuously conveyed in multiple groups of test empty cans includes: recording the image data of all rejected test empty cans through the visual inspection device; when there are test empty cans that have not been rejected, testing the test empty cans that have not been rejected again, and controlling the visual inspection device to record the image data of the test empty cans that have not been rejected; and using the image data of all rejected test empty cans and the image data of the test empty cans that have not been rejected as the test result of the continuous can rejection.

[0015] Furthermore, after the step of obtaining a database of can body visual inspection based on the test results of the single rejection test and the test results of the continuous can rejection test, the step also includes: using a visual inspection device to perform accuracy inspection on the continuously conveyed and spaced test empty cans and qualified can bodies.

[0016] Using the technical solution of the present invention, multiple groups of test empty cans are first produced. A visual inspection device is then used to perform a single rejection test on individual test empty cans within the multiple groups, obtaining a single rejection test result. The visual inspection device is then used to perform a continuous can rejection test on multiple test empty cans continuously conveyed within the multiple groups, obtaining a continuous can rejection test result. Finally, the test results of the single rejection test and the continuous can rejection test are combined to create a visual inspection database. Specifically, the test results of the single rejection test provide data support for the continuous can rejection test, thereby improving the accuracy of the continuous can rejection test. After combining the test results of the single rejection test and the continuous can rejection test to create a visual inspection database, the database is then compared with actual inspected empty cans, thereby improving inspection accuracy and avoiding missed or erroneous detections. Therefore, the technical solution of the present application effectively addresses the problem of high randomness and proneness to missed detections in related visual inspection techniques. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0018] Figure 1 A flowchart showing the overall steps of an embodiment of a method for building a database for visual inspection of tank bodies according to the present invention is shown;

[0019] Figure 2 Shown Figure 1 Specific flow chart of step S10 of the method for building a database for visual inspection of tank bodies;

[0020] Figure 3 Shown Figure 1 Specific flow chart of step S20 of the method for building a database for visual inspection of tank bodies;

[0021] Figure 4 Shown Figure 1 Specific flow chart of step S30 of the method for building a database for visual inspection of tank bodies;

[0022] Figure 5 Shown Figure 1 Schematic diagram of the structure of the device required for the database construction method of tank visual inspection.

[0023] The above drawings include the following reference numerals:

[0024] 10. Test empty cans; 20. Visual inspection device. DETAILED DESCRIPTION

[0025] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is in no way intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0027] Unless otherwise specifically stated, the relative arrangement of the parts and steps, the numerical expressions and the numerical values ​​set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The techniques, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods and equipment should be considered as part of the authorization specification. In all examples shown and discussed here, any specific values ​​should be interpreted as being merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that similar numbers and letters represent similar items in the following figures, and therefore, once an item is defined in one figure, it does not need to be further discussed in subsequent figures.

[0028] The visual inspection device 20 primarily utilizes grayscale image detection technology and checkered image detection technology. Grayscale image detection refers to the gradation of brightness levels in a displayed image, from brightest to darkest. The more grayscale levels, the more detailed the image. The software determines the pattern of grayscale value changes in the image through line scanning and boundary discrimination calculations, thereby quickly determining whether the image is abnormal. Checkered image detection primarily uses edge detection methods to determine whether the image presents regular square boundaries, and determines whether the image is abnormal by calculating pixel grayscale values.

[0029] Essentially, the visual inspection device 20 is a camera system that integrates optoelectronic technology, computer digital signal processing, and software analysis. When a product passes through the camera at a constant speed, the camera quickly captures the image. Once the complete image is captured, the software detection tool will check for abnormalities. However, during this process, the computer must test a large number of abnormal samples that simulate actual production conditions, and then continuously correct parameters such as grayscale values, recognition area shapes, and recognition areas until the accuracy is met. If the accuracy is too high, it will result in a high rate of product misjudgment, affecting production efficiency; if the accuracy is too low, abnormal products will not be detected, so the detection accuracy of the system is crucial.

[0030] Currently, there is no standardized verification method for various types of visual inspection devices 20 on the market. Generally, the accuracy of the equipment is continuously debugged through long-term exploration of production process data and experience. However, this design method lacks standardized simulation test samples, is highly random, inefficient, and has poor reproducibility. It cannot simulate the actual production status and cannot achieve accurate effectiveness testing and system accuracy correction. The entire debugging cycle is very long.

[0031] The technical solution of this embodiment first establishes a complete set of standard test samples to simulate the product characteristics under real production conditions, and then establishes a complete standardized testing method for different test samples. This testing method is widely applicable to the rejection effectiveness verification test of the visual inspection device 20 of various can types.

[0032] Specifically, if Figure 1 and Figure 5 As shown, in this embodiment, the method for building a database for tank visual inspection includes:

[0033] Step S10: obtaining multiple groups of test empty cans;

[0034] Step S20: using the visual inspection device 20 to perform a single rejection test on a single test empty can in the plurality of test empty cans to obtain a single rejection test result;

[0035] Step S30: using the visual inspection device 20 to perform a continuous can rejection test on the plurality of test empty cans continuously conveyed in the plurality of test empty can groups to obtain a continuous can rejection test result;

[0036] Step S40: obtaining a database of can body visual inspection according to the test results of the single rejection test and the test results of the continuous can rejection test.

[0037] Using the technical solution of this embodiment, multiple groups of test empty cans are first prepared to obtain multiple test groups of empty cans. Then, a visual inspection device 20 is used to perform a single rejection test on individual test empty cans from the multiple test groups, obtaining single rejection test results. Then, the visual inspection device 20 is used to perform a continuous can rejection test on multiple test empty cans continuously conveyed from the multiple test groups, obtaining continuous can rejection test results. Finally, the test results of the single rejection test and the continuous can rejection test are combined to form a visual inspection database. Specifically, the test results of the single rejection test provide data support for the continuous can rejection test, thereby improving the accuracy of the continuous can rejection test. After combining the test results of the single rejection test and the continuous can rejection test to form a visual inspection database, the database is compared with actual inspected empty cans, thereby improving inspection accuracy and avoiding missed or erroneous detections. Therefore, the technical solution of this embodiment effectively addresses the problem of high randomness and proneness to missed detections in related visual inspection techniques.

[0038] Furthermore, a single rejection test is performed first and then a continuous can rejection test is performed, so that machine learning of the visual inspection device 20 can be achieved, thereby improving the accuracy of the inspection.

[0039] like Figure 5 As shown, the test empty can shown in the figure is designated by serial number 10.

[0040] like Figure 1 and Figure 2 As shown, in this embodiment, the steps of obtaining multiple groups of test empty cans include:

[0041] Step S11: placing foreign matter inside the can to obtain a first set of test empty cans;

[0042] Step S12: squeezing and deforming the can bodies to obtain a second set of test empty cans;

[0043] Step S13: cutting the mouth or bottom edge of the can body to obtain a third group of test empty cans.

[0044] Specifically, foreign matter includes oil-based ink, paper scraps, and hair. The first group of test empty cans includes painting oil-based ink on the side wall of the can, painting oil-based ink on the bottom wall of the can, painting oil-based ink on the connection between the side wall and the bottom wall of the can, sticking paper scraps or hair on the side wall of the can, sticking paper scraps or hair on the bottom wall of the can, and sticking paper scraps or hair on the connection between the side wall and the bottom wall of the can. Oil-based ink is used to simulate small insects or dust. After squeezing the can and deforming it, the second group of test empty cans can be obtained. Specifically, the second group of test empty cans includes deformation of the side wall of the can, deformation of the opening of the can, and when the opening of the can is elliptical, the difference between the two radii of the ellipse is less than or equal to 1 mm. The third group of test empty cans includes shearing or folding at the edge of the mouth of the can (drooping lip, broken curling or curling), and shear notches on the side wall or bottom wall of the can. Specifically, the width of the vertical lip is less than or equal to 2.5 cm, the diameter of the notch is less than or equal to 0.5 cm, and the diameter of the curled edge is less than or equal to 0.5 cm.

[0045] like Figure 1 and Figure 3 As shown, in this embodiment, step S20: using the visual inspection device 20 to perform a single rejection test on a single test empty can in the multiple groups of test empty cans to obtain a single rejection test result includes:

[0046] Step S21: performing multiple tests on multiple test empty cans with different foreign matter positions in the first group of test empty cans, and performing multiple tests on each test empty can in the first group of test empty cans;

[0047] Step S22: performing multiple tests on the multiple test empty cans of different shapes in the second group of test empty cans, and performing multiple tests on each test empty can in the second group of test empty cans;

[0048] Step S23: performing multiple tests on the multiple test empty cans with different cutout positions in the third group of test empty cans, and performing multiple tests on each test empty can in the third group of test empty cans.

[0049] The above setting can better simulate the real test. Specifically, each test needs to rotate the empty can 30 degrees clockwise along the axis of the test. If all the empty cans are rejected normally, it is judged as qualified. Otherwise, it is considered as failed.

[0050] like Figure 1 and Figure 3 As shown, in this embodiment, step S21: the step of performing multiple tests on each test empty can in the first group of test empty cans includes:

[0051] Each test is performed by rotating the empty can to the first preset angle or by turning it upside down;

[0052] Step S22: Perform multiple tests on each test empty can in the second group of test empty cans:

[0053] Each test is performed by rotating the empty can to the first preset angle or by turning it upside down;

[0054] Step S23: Perform multiple tests on each test empty can in the third group of test empty cans:

[0055] Each test is performed by rotating the empty can to a first preset angle or by turning it upside down.

[0056] The above settings can accurately simulate the real situation, which can provide data support for subsequent real situations. The number of multiple tests includes at least 30 times.

[0057] like Figure 1 and Figure 4 As shown, in this embodiment, the steps of performing a continuous can rejection test on a plurality of continuously conveyed test empty cans in a plurality of groups of test empty cans using the visual inspection device 20 to obtain a continuous can rejection test result include:

[0058] Step S31: Continuously testing the first group of test empty cans using the visual inspection device 20;

[0059] Step S32: continuously testing the second group of test empty cans using the visual inspection device 20;

[0060] Step S33: Continuously test the third group of test empty cans using the visual inspection device 20 .

[0061] The above-mentioned setting increases the requirements for the testing accuracy of the visual inspection device 20, requiring the detection accuracy of the visual inspection device 20 to be more precise.

[0062] like Figures 1 to 5 As shown, in this embodiment, the steps of performing a continuous can rejection test on a plurality of continuously conveyed test empty cans in a plurality of groups of test empty cans using the visual inspection device 20 to obtain a continuous can rejection test result include:

[0063] When multiple sets of test empty cans are continuously tested using the visual inspection device 20, the second test empty can of two adjacent test empty cans is rotated relative to the first test empty can by a second preset angle. This arrangement places higher demands on the detection accuracy of the visual inspection device 20, thereby ensuring detection accuracy.

[0064] like Figures 1 to 5 As shown, in this embodiment, the first preset angle is 30° and the second preset angle is 60°. The above angles can be realistically simulated. Of course, the above first preset angle can also be other angles, specifically between 10° and 90°. The second preset angle can be between 10° and 90°.

[0065] like Figures 1 to 5 As shown, in this embodiment, when performing continuous testing using multiple groups of test empty cans and the visual inspection device 20, the step of rotating the second test empty can of two adjacent test empty cans by a second preset angle relative to the first test empty can includes forming a test set of six test empty cans, with the first test empty can in each test set having a different rotation angle. This configuration allows for more realistic simulation, thereby enriching the database.

[0066] like Figures 1 to 5 As shown, in this embodiment, the step of using the visual inspection device 20 to perform a single rejection test on a single test empty can in a plurality of test groups of empty cans to obtain a single rejection database includes: recording the image data of all rejected test empty cans by the visual inspection device 20, and obtaining a single rejection test result based on the image data of all rejected test empty cans; and the step of using the visual inspection device 20 to perform a continuous can rejection test on a plurality of test empty cans continuously conveyed in the plurality of test groups of empty cans to obtain a continuous can rejection test result includes: recording the image data of all rejected test empty cans by the visual inspection device 20, and obtaining a continuous can rejection test result based on the image data of all rejected test empty cans. The above-mentioned configuration increases the data volume of the single rejection test results and the continuous can rejection test results, thereby providing a larger sample size for subsequent actual testing.

[0067] like Figures 1 to 5 As shown, in this embodiment, the step of using the visual inspection device 20 to perform a single rejection test on a single test empty can in multiple groups of test empty cans includes: recording the image data of all rejected test empty cans through the visual inspection device 20; when there are test empty cans that are not rejected, the test empty cans that are not rejected are tested again, and the visual inspection device 20 is controlled to record the image data of the test empty cans that are not rejected; recording the image data of all rejected test empty cans and the image data of the test empty cans that are not rejected as the test results of the single rejection test; the step of using the visual inspection device 20 to perform a continuous can rejection test on multiple test empty cans that are continuously conveyed in multiple groups of test empty cans includes: recording the image data of all rejected test empty cans through the visual inspection device 20; when there are test empty cans that are not rejected, the test empty cans that are not rejected are tested again, and the visual inspection device 20 is controlled to record the image data of the test empty cans that are not rejected; and using the image data of all rejected test empty cans and the image data of the test empty cans that are not rejected as the test results of the continuous can rejection. The above-mentioned setting can record actively and make the database of the visual inspection device 20 more accurate.

[0068] like Figures 1 to 5As shown, in this embodiment, after the step of obtaining a database for can body visual inspection based on the test results of the single rejection test and the continuous can rejection test, the method further includes: using the visual inspection device 20 to perform accuracy inspection on the continuously conveyed and spaced empty test cans and qualified can bodies. The above-mentioned configuration enables verification, and after verification, it can be checked whether the visual inspection device 20 can perform inspections quickly and accurately.

[0069] Specifically, the detection rate = (number of rejections / total number of test samples) × 100%; the false kick rate = (number of false kicks / total number of test samples) × 100%. In terms of the detection rate and false kick rate, the specified standardized test samples are used to conduct intermittent rejection tests on samples (intermittent, that is, abnormal cans and normal cans are placed alternately or with one can position between them, and abnormal cans shall not be placed for more than 2 cans continuously). Each type of test is conducted at least 30 times (each time rotating 60 degrees clockwise along the axis). If the detection rate of each type reaches 100%, it is judged to be qualified, otherwise it is deemed to have failed the test. The rejection test is conducted under simulated or normal production conditions, and at least 10,000 tests are conducted (continuously). If the false kick rate is ≤0.05%, it is judged to be qualified, otherwise it is deemed to have failed the test.

[0070] like Figures 1 to 5 As shown, the technical solution of this embodiment establishes a complete set of methods for producing standard test samples, which can fully simulate the product characteristics under real production conditions and are widely applicable to different types of cans. A complete set of standardized test methods are established for different test samples. The combination of standard test samples and standardized test methods can quickly complete system validity verification and accuracy calibration, improving timeliness, greatly optimizing system detection accuracy, improving detection rate, and reducing false kick rate.

[0071] In the description of the present invention, it should be understood that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "horizontal, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific direction or be constructed and operated in a specific direction. Therefore, they cannot be understood as limiting the scope of protection of the present invention; the directional words "inside and outside" refer to the inside and outside relative to the outline of each component itself.

[0072] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used herein to describe the spatial positional relationship of a device or feature to other devices or features as shown in the figures. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figures. For example, if the device in the drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0073] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.

[0074] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for building a database for tank visual inspection, characterized in that: include: Get multiple groups of test empty cans; Using a visual inspection device (20) to perform a single rejection test on a single test empty can in the plurality of test empty cans to obtain a single rejection test result; Using the visual inspection device (20) to perform a continuous can rejection test on a plurality of the continuously conveyed test empty cans in a plurality of groups of the test empty cans to obtain a continuous can rejection test result; Obtaining a database for can visual inspection based on the test results of the single rejection test and the continuous can rejection test; When continuous testing is performed using multiple groups of the test empty cans and the visual inspection device (20), the step of rotating the latter of two adjacent test empty cans relative to the former test empty can by a second preset angle comprises: The six test empty cans are formed into a test set, wherein the rotation angle of the first test empty can in each test set is different; The steps of performing a single rejection test on a single test empty can in the plurality of test empty cans using a visual inspection device (20) include: Recording image data of all rejected test empty cans by the visual inspection device (20); When there are the test empty cans that have not been rejected, the test empty cans that have not been rejected are tested again, and the visual inspection device (20) is controlled to record image data of the test empty cans that have not been rejected; Recording the image data of all rejected test empty cans and the image data of the test empty cans that are not rejected as the test results of the single rejection test; The steps of using the visual inspection device (20) to perform a continuous can rejection test on a plurality of the continuously conveyed test empty cans in a plurality of groups of the test empty cans include: Recording image data of all rejected test empty cans by the visual inspection device (20); When there are the test empty cans that have not been rejected, the test empty cans that have not been rejected are tested again, and the visual inspection device (20) is controlled to record image data of the test empty cans that have not been rejected; The image data of all rejected test empty cans and the image data of the test empty cans that are not rejected are used as the test results of continuous can rejection.

2. The method for building a database for tank visual inspection according to claim 1, characterized in that: The steps to obtain multiple sets of empty test cans include: placing foreign objects inside the can body to obtain a first group of test empty cans; The can bodies are squeezed and deformed to obtain the second group of empty test cans; The mouth or bottom edge of the can body is cut to obtain the third group of test empty cans.

3. The method for building a database for tank visual inspection according to claim 2, characterized in that: The steps of using a visual inspection device (20) to perform a single rejection test on a single test empty can in the plurality of test empty cans to obtain a single rejection test result include: Performing multiple tests on the plurality of test empty cans in the first group of test empty cans having different foreign matter positions, and performing multiple tests on each of the test empty cans in the first group of test empty cans; performing multiple tests on the plurality of test empty cans of different shapes in the second group of test empty cans, and performing multiple tests on each of the test empty cans in the second group of test empty cans; Multiple tests are performed on the plurality of test empty cans with different cutout positions in the third group of test empty cans, and multiple tests are performed on each of the test empty cans in the third group of test empty cans.

4. The method for building a database for visual inspection of tank bodies according to claim 3, characterized in that: The step of performing multiple tests on each of the test empty cans in the first group of test empty cans comprises: Each test is performed by rotating the test empty can by a first preset angle or inverting the test empty can; Perform multiple tests on each of the test empty cans in the second group of test empty cans: Each test is performed by rotating the test empty can by a first preset angle or inverting the test empty can; Perform multiple tests on each of the test empty cans in the third group of test empty cans: Each test is performed by rotating the test empty can by a first preset angle or by turning it upside down.

5. The method for building a database for visual inspection of tanks according to claim 4, characterized in that: The steps of using the visual inspection device (20) to perform a continuous can rejection test on a plurality of the continuously conveyed test empty cans in a plurality of groups of the test empty cans to obtain a continuous can rejection test result include: Continuously testing the first group of test empty cans using the visual inspection device (20); Continuously testing the second group of test empty cans using the visual inspection device (20); The third group of test empty cans is continuously tested by the visual inspection device (20).

6. The method for building a database for visual inspection of tanks according to claim 5, characterized in that: The steps of using the visual inspection device (20) to perform a continuous can rejection test on a plurality of the continuously conveyed test empty cans in a plurality of groups of the test empty cans to obtain a continuous can rejection test result include: When continuous testing is performed using multiple groups of the test empty cans and the visual inspection device (20), the latter of two adjacent test empty cans is rotated by a second preset angle relative to the former test empty can.

7. The method for building a database for visual inspection of tank bodies according to claim 6, characterized in that: The first preset angle is 30°, and the second preset angle is 60°.

8. The method for building a database for visual inspection of tanks according to claim 1, characterized in that: The steps of using a visual inspection device (20) to perform a single rejection test on a single test empty can in the plurality of test empty can groups to obtain a single rejection database include: Recording the image data of all rejected test empty cans by the visual inspection device (20), and obtaining the single rejection test result based on the image data of all rejected test empty cans; The steps of using the visual inspection device (20) to perform a continuous can rejection test on a plurality of the continuously conveyed test empty cans in a plurality of groups of the test empty cans to obtain a continuous can rejection test result include: The visual inspection device (20) records the image data of all rejected test empty cans, and obtains a test result of continuous can rejection based on the image data of all rejected test empty cans.

9. The method for building a database for visual inspection of tanks according to claim 1, characterized in that: After the step of obtaining a database for can visual inspection based on the test results of the single rejection test and the test results of the continuous can rejection test, the following steps are further included: The visual inspection device (20) is used to perform accuracy inspection on the test empty cans and qualified can bodies that are continuously conveyed and spaced apart.

Images