Aluminum foil meal box appearance detection system

Abstract

The invention discloses an aluminum foil meal box appearance detection system, which comprises a detection line transmission mechanism, one end of the detection line transmission mechanism is provided with a head end covering detection mechanism and a double-end sorting mechanism, and the head end covering detection mechanism is used for detecting the defect performance of an aluminum foil meal box; the detection end of the detection surface lifting turnover mechanism extends to the inner side of the aluminum foil meal box in a lifting state and then moves along the seam gap of the aluminum foil meal box to acquire a local image, and after the image is acquired, the appearance acquisition image analyzer compares a pre-recorded qualified contrast image with a current image to analyze whether the inner gap has defects or not. When obvious inner gap defects exist, defective products are sorted and picked out by the double-end sorting mechanism, and the aluminum foil meal boxes without the obvious inner gap defects are labeled and calibrated by the detection surface lifting and overturning mechanism, so that the appearance defects such as stains, spots, roll marks and joints of the aluminum foil meal boxes are quickly detected, the appearance detection accuracy is ensured, and the detection efficiency is improved. And the manual approval operation amount is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of appearance detection, and in particular to an appearance detection system for an aluminum foil lunch box. Background Art

[0002] In the production process of aluminum foil, various defects often occur on the surface of aluminum foil due to the influence of various factors such as raw materials, production sites and production technology. The defects are mainly damage defects such as holes, appearance defects such as stains and spots, use defects such as pits and bubbles, and abnormal defects at the connection caused by roller marks and seams. In the traditional detection system, the infrared detection method is mainly used, which uses high-frequency induction coils to generate induced current on the surface of the object. The skin effect of high-frequency induction is used to make the current inside the object flow only at a distance from the surface. That is, if there are cracks or other defects on the outer surface of the detected object, the defect position will generate abnormal energy loss, thereby quickly identifying the defect position. This detection method relies on generating induced current on the surface of the object. When detecting non-morphological appearance defects such as stains, spots, roller marks and seams, as well as abnormal defects on abnormal connection surfaces, the induced current response deviation is small, the detection accuracy is not ideal, and the detection conditions required for implementation are too complicated. Therefore, in order to ensure the accuracy of stain, spot, roller mark and seam detection, we propose an aluminum foil lunch box appearance detection system. Summary of the invention

[0003] The main purpose of the invention is to provide an aluminum foil lunch box appearance detection system.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] An aluminum foil lunch box appearance inspection system comprises a detection line transmission mechanism, one end of which is provided with a head end coverage detection mechanism and a double end sorting mechanism, the head end coverage detection mechanism is used to detect the defect performance of the aluminum foil lunch box, and sends a sorting signal to the double end sorting mechanism after the detection, and the aluminum foil lunch boxes with morphological defects and the aluminum foil lunch boxes of unrecognizable types are dispersedly transported by the double end sorting mechanism, wherein the double end sorting mechanism transports the aluminum foil lunch boxes with morphological defects to a waste area, and transports the aluminum foil lunch boxes of unrecognizable defect types to a secondary operation platform, the secondary operation platform is installed at the transmission tail end of the detection line transmission mechanism, and the secondary operation platform is provided with a detection surface lifting and flipping mechanism, a matrix set image acquisition module and an appearance acquisition image analyzer;

[0006] After the double-end sorting mechanism transports the aluminum foil lunch box whose defect type cannot be identified to the secondary working platform, the matrix set image acquisition module lifts the aluminum foil lunch box, and the detection end of the detection surface lifting and flipping mechanism extends to the inner side of the lifted aluminum foil lunch box to obtain its surface-level structural image and perform low-grayscale debugging processing. After the acquisition, the detection surface lifting and flipping mechanism continues to move along the seam gap of the aluminum foil lunch box to collect local images, and the appearance acquisition image analyzer pre-records a qualified seam performance image. After the detection surface lifting and flipping mechanism obtains the image performance after the motion detection, the appearance acquisition image analyzer compares the pre-recorded qualified control image with the current image performance to analyze its appearance defects.

[0007] A further improvement of the present invention is that the head-end coverage detection mechanism includes a thick-walled surface ultrasonic detection probe, a vertical base surface ultrasonic detection probe and an identification signal processing unit; the thick-walled surface ultrasonic detection probe and the vertical base surface ultrasonic detection probe are fixedly connected to the detection line transmission mechanism, and the detection and collection ends of the thick-walled surface ultrasonic detection probe and the vertical base surface ultrasonic detection probe are both extended to the inside of the detection line transmission mechanism, and the signal transmission ends of the thick-walled surface ultrasonic detection probe and the vertical base surface ultrasonic detection probe are both signal-connected to the identification signal processing unit; the thick-walled surface ultrasonic detection probe and the vertical base surface ultrasonic detection probe respectively identify whether there are morphological defects in the aluminum foil lunch box after it is conveyed.

[0008] A further improvement of the present invention is that the morphological defects detected by the thick-wall surface ultrasonic detection probe and the vertical base surface ultrasonic detection probe include any one of pits, bubbles and large-scale folds.

[0009] A further improvement of the present invention is that the double-end sorting mechanism includes a planar lifting claw and a distribution robot arm assembly, and the distribution robot arm assembly is used to drive the planar lifting claw to move at multiple angles along the fixed surface for transporting the aluminum foil lunch box, so that the planar lifting claw can drag the aluminum foil lunch box flatly to the secondary working platform under non-pressure contact.

[0010] A further improvement of the present invention is that the matrix set image acquisition module includes a fixed-track displacement actuator, a small target matrix probe and a shading surface angle adjustment actuator, the fixed-track displacement actuator is used to obtain the surface-level structural image of the aluminum foil lunch box along the vertical plane, the shading surface angle adjustment actuator is used to drive the small target matrix probe to move in a circular motion along the seam gap of the aluminum foil lunch box, and the local image of the seam gap of the aluminum foil lunch box is obtained by the small target matrix probe, and the small target matrix probe and the shading surface angle adjustment actuator are both connected to the signal of the appearance acquisition image analyzer.

[0011] A further improvement of the present invention is that a method for using an aluminum foil lunch box appearance detection system comprises the following steps:

[0012] Step S1, the finished aluminum foil lunch box is transported to the head end coverage detection mechanism through the detection line transmission mechanism, and after the head end coverage detection mechanism performs ultrasonic detection, the aluminum foil lunch box identified as having morphological defects is sorted and removed through the double-end sorting mechanism, and the aluminum foil lunch box that cannot be identified is sent to the secondary operation platform;

[0013] Step S, after the secondary operation platform receives the type of defect that cannot be identified, the detection surface lifting and flipping mechanism lifts the aluminum foil lunch box, and the matrix set image acquisition module obtains the low grayscale surface level structure image of the aluminum foil lunch box, and the appearance acquisition image analyzer 7 compares and analyzes the pre-recorded qualified control image with the current image performance to obtain whether the aluminum foil lunch box plane has appearance defects. If appearance defects occur, the double-end sorting mechanism sorts out the defective products, and the aluminum foil lunch boxes without obvious defects are inspected for internal gap defects by the matrix set image acquisition module;

[0014] Step S, when detecting the inner gap defect, first extend the detection end of the matrix set image acquisition module to the inner side of the aluminum foil lunch box in the lifted state, and then move along the seam gap of the aluminum foil lunch box to collect a local image. After acquiring the image, the appearance acquisition image analyzer 7 compares the pre-recorded qualified control image with the current image performance to analyze whether there is a defect in the inner gap. When there is an obvious inner gap defect, the defective products are sorted out by the double-end sorting mechanism, and the aluminum foil lunch boxes without obvious inner gap defects are calibrated by affixing labels by the matrix set image acquisition module and output to the manual verification area.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: after ultrasonic detection by the first-end covering detection mechanism, the aluminum foil lunch boxes identified as having morphological defects are sorted and removed by the double-end sorting mechanism, and the unrecognizable types are sent to the secondary operating platform for internal gap defect detection on the secondary operating platform, that is, the detection end of the detection surface lifting and flipping mechanism is extended to the inner side of the aluminum foil lunch box in the lifting state, and then moves along the seam gap of the aluminum foil lunch box to collect local images. After the image is obtained, the appearance acquisition image analyzer compares the pre-recorded qualified control image with the current image performance to analyze whether there is a defect in the internal gap. When there is an obvious internal gap defect, the defective product is sorted and picked out by the double-end sorting mechanism, and the aluminum foil lunch box without obvious internal gap defects is calibrated by affixing labels by the detection surface lifting and flipping mechanism, thereby quickly detecting stains, spots, roller marks and seam-related appearance defects of the aluminum foil lunch box, ensuring the accuracy of appearance detection while reducing the amount of manual verification operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the structure of an aluminum foil lunch box appearance detection system of the present invention. Figure 1 .

[0017] Figure 2This is a schematic diagram of the structure of an aluminum foil lunch box appearance detection system of the present invention. Figure 2 .

[0018] Figure 3 The present invention is a flow chart of an aluminum foil lunch box appearance detection system.

[0019] In the figure: 1. Detection line transmission mechanism; 2. Head-end coverage detection mechanism; 21. Thick-wall ultrasonic detection probe; 22. Vertical base ultrasonic detection probe; 23. Identification signal processing unit; 3. Double-end sorting mechanism; 31. Plane lifting claw; 32. Distribution robot arm assembly; 4. Secondary working platform; 5. Detection surface lifting and flipping mechanism; 6. Matrix image acquisition module; 61. Track-fixed displacement actuator; 62. Small target matrix probe; 63. Shading surface angle adjustment actuator; 7. Appearance acquisition image analyzer. DETAILED DESCRIPTION

[0020] The present invention is further described below in conjunction with specific embodiments, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limitations on this patent. In order to better illustrate the specific embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the accompanying drawings may be omitted. Based on the specific embodiments of the present invention, all other specific embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] Example

[0022] See also Figure 1-Figure 3 A system for detecting the appearance of an aluminum foil lunch box comprises a detection line transmission mechanism 1, wherein one end of the detection line transmission mechanism 1 is provided with a head end coverage detection mechanism 2 and a double end sorting mechanism 3, wherein the head end coverage detection mechanism 2 is used to detect the defect performance of the aluminum foil lunch box, and sends a sorting signal to the double end sorting mechanism 3 after the detection, and the aluminum foil lunch boxes with morphological defects and the aluminum foil lunch boxes of unrecognizable types are dispersedly transported through the double end sorting mechanism 3, wherein the double end sorting mechanism 3 transports the aluminum foil lunch boxes with morphological defects to a waste area, and transports the aluminum foil lunch boxes of unrecognizable defect types to a secondary operating platform 4, wherein the secondary operating platform 4 is installed at the transmission tail end of the detection line transmission mechanism 1, and the secondary operating platform 4 is provided with a detection surface lifting and flipping mechanism 5, a matrix set image acquisition module 6 and an appearance acquisition image analyzer 7.

[0023] In the present embodiment, in order to solve the defects that the detection accuracy of the traditional detection method for non-morphological appearance defects such as stains, spots, roller marks and seams and abnormal defects on abnormal connection surfaces is not ideal, and the detection conditions required for implementation are too complicated, the present invention proposes an aluminum foil lunch box appearance detection system, wherein, since the improvement of the present invention lies in the detection of non-morphological appearance defects such as stains, spots, roller marks and seams, no technical improvement is made to the detection line transmission mechanism 1 and the double-end sorting mechanism 3, and therefore the technical principle of the detection line transmission mechanism 1 to transmit the aluminum foil lunch box and the double-end sorting mechanism 3 to sort out the aluminum foil lunch box after the head end covers the detection mechanism 2 The detection mechanism 2 is not described in detail in the present invention.

[0024] In an optional implementation of the present embodiment, after the double-end sorting mechanism 3 transports the aluminum foil lunch box whose defect type cannot be identified to the secondary working platform 4, the matrix set image acquisition module 6 lifts the aluminum foil lunch box, and the detection end of the detection surface lifting and flipping mechanism 5 extends to the inner side of the lifted aluminum foil lunch box to obtain its surface level structure image and perform low grayscale debugging processing. After the acquisition, the detection surface lifting and flipping mechanism 5 continues to move along the seam gap of the aluminum foil lunch box to collect local images, and the appearance acquisition image analyzer 7 pre-records a qualified seam performance image. After the detection surface lifting and flipping mechanism 5 obtains the image performance after the motion detection, the appearance acquisition image analyzer 7 compares the pre-recorded qualified control image with the current image performance to analyze its appearance defects.

[0025] In this embodiment, the present invention performs appearance defect detection of stains, spots, roller marks and seams in the outer aluminum foil lunch box by setting up a secondary working platform 4, a detection surface lifting and flipping mechanism 5, a matrix set image acquisition module 6 and an appearance acquisition image analyzer 7. The plane lifting claw head 31 is driven by the distribution robot arm assembly 32 to flatly drag the aluminum foil lunch box whose defect type cannot be identified to the secondary working platform 4. After the transportation is completed, the detection surface lifting and flipping mechanism 5 realizes support for the lunch box during inspection by supporting the four corners of the lunch box with a plane.

[0026] In an optional implementation of the present embodiment, the head-end coverage detection mechanism 2 includes a thick-wall surface ultrasonic detection probe 21, a vertical base surface ultrasonic detection probe 22 and an identification signal processing unit 23. The thick-wall surface ultrasonic detection probe 21 and the vertical base surface ultrasonic detection probe 22 are fixedly connected to the detection line transmission mechanism 1, and the detection and collection ends of the thick-wall surface ultrasonic detection probe 21 and the vertical base surface ultrasonic detection probe 22 are both extended to the inner side of the detection line transmission mechanism 1, and the signal transmission ends of the thick-wall surface ultrasonic detection probe 21 and the vertical base surface ultrasonic detection probe 22 are both connected to the identification signal processing unit 23 by signal. The thick-wall surface ultrasonic detection probe 21 and the vertical base surface ultrasonic detection probe 22 respectively identify whether there are morphological defects in the aluminum foil lunch box after it is conveyed.

[0027] In this embodiment, the thick-walled ultrasonic detection probe 21 and the vertical base ultrasonic detection probe 22 can be extended to the inside of the aluminum foil lunch box by the controllable mechanical arm during detection, wherein the thick-walled ultrasonic detection probe 21 corresponds to the inner and outer wall support surfaces of the lunch box for detection, and the vertical base ultrasonic detection probe 22 corresponds to the inner and outer sides of the bottom surface of the lunch box for detection. After the thick-walled ultrasonic detection probe 21 and the vertical base ultrasonic detection probe 22 complete the detection, the identification signal processing unit 23 sends the corresponding sorting processing signal to the double-end sorting mechanism 3, and the lunch boxes identified as morphological defects are removed by the driving of the distribution mechanical arm assembly 32 and the plane lifting claw head 31, while the lunch boxes with shallow defects and unrecognizable morphological defect types are grasped by the distribution mechanical arm assembly 32 and driven by the plane lifting claw head 31 to be transmitted to the secondary working platform 4, and the matrix set image acquisition module 6 set in the secondary working platform 4 adopts the small target matrix set for detection.

[0028] In an optional implementation of this embodiment, the morphological defects detected by the thick wall ultrasonic detection probe 21 and the vertical base surface ultrasonic detection probe 22 include any one of pits, bubbles and large-scale folds.

[0029] In an optional implementation of the present embodiment, the double-end sorting mechanism 3 includes a planar lifting claw head 31 and a distribution robot arm assembly 32. The distribution robot arm assembly 32 is used to drive the planar lifting claw head 31 to move at multiple angles along the fixed surface for transporting the aluminum foil lunch box, so that the planar lifting claw head 31 can drag the aluminum foil lunch box flatly to the secondary working platform 4 without pressure contact.

[0030] In an optional implementation of the present embodiment, the matrix set image acquisition module 6 includes a track-fixed displacement actuator 61, a small target matrix probe 62 and a shading surface angle adjustment actuator 63. The track-fixed displacement actuator 61 is used to obtain the surface-level structural image of the aluminum foil lunch box along the vertical plane, and the shading surface angle adjustment actuator 63 is used to drive the small target matrix probe 62 to move in a circular motion along the seam gap of the aluminum foil lunch box, so as to obtain a local image of the seam gap of the aluminum foil lunch box through the small target matrix probe 62. The small target matrix probe 62 and the shading surface angle adjustment actuator 63 are both connected to the signal of the appearance acquisition image analyzer 7.

[0031] In this embodiment, the shading surface angle adjustment actuator 63 is installed integrally with the small target matrix probe 62 and moves synchronously with the drive of the track displacement actuator 61. The shading surface angle adjustment actuator 63 is used to fill in the imaging surface of the small target matrix probe 62 during collection, ensuring that when stains, spots, roller marks and seams exist, the small target matrix probe 62 can accurately extract the image of the position, thereby quickly realizing the verification of appearance defects. When the small target matrix probe 62 performs preliminary inspection, the small target matrix probe 62 is raised to a high position and vertically covers the entire aluminum foil lunch box in a large range, thereby obtaining the overall surface-level structural image of the aluminum foil lunch box. The overall surface-level structural image of the aluminum foil lunch box is preliminarily screened by the image processor to obtain whether there are obvious stains, spots, roller marks and seams on the aluminum foil lunch box. If there are obvious stains, spots, roller marks and seams, the aluminum foil lunch box is fixed by the detection surface lifting and flipping mechanism 5, and the distribution mechanical arm assembly 32 is driven by the plane lifting claw 31 to lift and pick out the aluminum foil lunch box supported by the detection surface lifting and flipping mechanism 5. When the shading surface angle adjustment actuator 63 completes the appearance defect verification, if the current aluminum foil lunch box does not have obvious difference defects, the small target matrix probe 62 is driven by the fixed track displacement actuator 61 to move along the gap position of the inner edge corner seam of the aluminum foil lunch box, and the small target matrix probe 62 obtains a local image at the gap of the seam of the aluminum foil lunch box, and then the local image is used to verify again whether there is a more obvious appearance defect surface.

[0032] The working principle of the present invention comprises the following steps:

[0033] Step S1, the finished aluminum foil lunch box is transported to the head end coverage detection mechanism 2 through the detection line transmission mechanism 1, and after the head end coverage detection mechanism 2 performs ultrasonic detection, the aluminum foil lunch box identified as having morphological defects is sorted and removed through the double-end sorting mechanism 3, and the aluminum foil lunch box that cannot be identified is sent to the secondary operation platform 4;

[0034] Step S2, after the secondary operation platform 4 receives the type of defect that cannot be identified, the detection surface lifting and flipping mechanism 5 lifts the aluminum foil lunch box, and the matrix set image acquisition module 6 obtains the low grayscale surface level structure image of the aluminum foil lunch box, and the appearance acquisition image analyzer 7 compares and analyzes the pre-recorded qualified control image with the current image performance to obtain whether the aluminum foil lunch box plane has appearance defects. If appearance defects occur, the double-end sorting mechanism 3 sorts out the defective products, and the aluminum foil lunch boxes without obvious defects are inspected for internal gap defects by the matrix set image acquisition module 6;

[0035] Step S3, when detecting the inner gap defect, first extend the detection end of the matrix set image acquisition module 6 to the inner side of the aluminum foil lunch box in the lifted state, and then move along the seam gap of the aluminum foil lunch box to collect a local image. After acquiring the image, the appearance acquisition image analyzer 7 compares the pre-recorded qualified control image with the current image performance to analyze whether there is a defect in the inner gap. When there is an obvious inner gap defect, the double-end sorting mechanism 3 sorts out the defective products, and the aluminum foil lunch boxes without obvious inner gap defects are labeled and calibrated by the matrix set image acquisition module 6 and output to the manual verification area.

Claims

1. An aluminum foil lunch box appearance inspection system, comprising a detection line transmission mechanism (1), characterized in that: One end of the detection line transmission mechanism (1) is provided with a head end coverage detection mechanism (2) and a double end sorting mechanism (3), wherein the head end coverage detection mechanism (2) is used to detect the defect manifestation of the aluminum foil lunch box and send a sorting signal to the double end sorting mechanism (3) after the detection, and the aluminum foil lunch boxes with morphological defects and the aluminum foil lunch boxes of unidentifiable types are dispersedly transported by the double end sorting mechanism (3), wherein the double end sorting mechanism (3) transports the aluminum foil lunch boxes with morphological defects to a waste area and transports the aluminum foil lunch boxes of unidentifiable defect types to a secondary operation platform (4), wherein the secondary operation platform (4) is installed at the transmission tail end of the detection line transmission mechanism (1), and the secondary operation platform (4) is provided with a detection surface lifting and flipping mechanism (5), a matrix set image acquisition module (6) and an appearance acquisition image analyzer (7); After the double-end sorting mechanism (3) conveys the aluminum foil lunch box whose defect type cannot be identified to the secondary operation platform (4), the matrix set image acquisition module (6) lifts the aluminum foil lunch box, and the detection end of the detection surface lifting and flipping mechanism (5) extends to the inner side of the aluminum foil lunch box in the lifted state to obtain its surface level structure image and perform low grayscale debugging processing. After the acquisition, the detection surface lifting and flipping mechanism (5) continues to move along the gap of the aluminum foil lunch box to collect local images, and the appearance acquisition image analyzer (7) pre-records a qualified seam performance image. After the detection surface lifting and flipping mechanism (5) obtains the image performance after the motion detection, the appearance acquisition image analyzer (7) compares the pre-recorded qualified control image with the current image performance to analyze its appearance defects.

2. The aluminum foil lunch box appearance detection system according to claim 1, characterized in that: The head end coverage detection mechanism (2) comprises a thick wall surface ultrasonic detection probe (21), a vertical base surface ultrasonic detection probe (22) and an identification signal processing unit (23); the thick wall surface ultrasonic detection probe (21) and the vertical base surface ultrasonic detection probe (22) are fixedly connected to the detection line transmission mechanism (1); the detection and collection ends of the thick wall surface ultrasonic detection probe (21) and the vertical base surface ultrasonic detection probe (22) are both extended to the inside of the detection line transmission mechanism (1); the signal transmission ends of the thick wall surface ultrasonic detection probe (21) and the vertical base surface ultrasonic detection probe (22) are both signal-connected to the identification signal processing unit (23); the thick wall surface ultrasonic detection probe (21) and the vertical base surface ultrasonic detection probe (22) respectively identify whether the aluminum foil lunch box has morphological defects after being conveyed.

3. The aluminum foil lunch box appearance detection system according to claim 2, characterized in that: The morphological defects detected by the thick-wall surface ultrasonic detection probe (21) and the vertical base surface ultrasonic detection probe (22) include any one of pits, bubbles and large-scale folds.

4. The aluminum foil lunch box appearance detection system according to claim 1, characterized in that: The double-end sorting mechanism (3) comprises a planar lifting claw head (31) and a dispensing mechanical arm assembly (32), wherein the dispensing mechanical arm assembly (32) is used to drive the planar lifting claw head (31) to move at multiple angles along the fixed surface for conveying the aluminum foil lunch box, so that the planar lifting claw head (31) can drag the aluminum foil lunch box flatly to the secondary working platform (4) without pressure contact.

5. The aluminum foil lunch box appearance detection system according to claim 1, characterized in that: The matrix image acquisition module (6) comprises a track-fixed displacement actuator (61), a small target matrix probe (62) and a shading surface angle adjustment actuator (63). The track-fixed displacement actuator (61) is used to obtain a surface-level structural image of the aluminum foil lunch box along a vertical plane. The shading surface angle adjustment actuator (63) is used to drive the small target matrix probe (62) to move in a circular manner along a seam gap of the aluminum foil lunch box, and obtain a local image of the seam gap of the aluminum foil lunch box through the small target matrix probe (62). The small target matrix probe (62) and the shading surface angle adjustment actuator (63) are both connected to the appearance acquisition image analyzer (7) by signal.

6. The steps of using the aluminum foil lunch box appearance detection system according to any one of claims 1 to 5 are characterized in that: The following steps are involved: Step S1, the finished aluminum foil lunch box is transported to the head end coverage detection mechanism (2) through the detection line transmission mechanism (1), and after the head end coverage detection mechanism (2) performs ultrasonic detection, the aluminum foil lunch boxes identified as having morphological defects are sorted and removed through the double-end sorting mechanism (3), and the aluminum foil lunch boxes that cannot be identified are sent to the secondary operation platform (4); Step S2, after the secondary operation platform (4) receives the type of defect that cannot be identified, the detection surface lifting and flipping mechanism (5) lifts the aluminum foil lunch box, and the matrix set image acquisition module (6) obtains the low grayscale surface level structure image of the aluminum foil lunch box, and the appearance acquisition image analyzer (7) compares and analyzes the pre-recorded qualified control image with the current image performance to obtain whether the aluminum foil lunch box surface has appearance defects. If appearance defects occur, the double-end sorting mechanism (3) sorts out the defective products, and the aluminum foil lunch boxes without obvious defects are inspected for internal gap defects by the matrix set image acquisition module (6); Step S3, when detecting the inner gap defect, firstly, the detection end of the matrix set image acquisition module (6) is extended to the inner side of the aluminum foil lunch box in the lifted state, and then the detection end is moved along the gap of the aluminum foil lunch box to collect a local image. After the image is obtained, the appearance acquisition image analyzer (7) compares the pre-recorded qualified control image with the current image to analyze whether there is a defect in the inner gap. When there is an obvious inner gap defect, the double-end sorting mechanism (3) sorts out the defective products, and the aluminum foil lunch boxes without obvious inner gap defects are labeled and calibrated by the matrix set image acquisition module (6) and output to the manual verification area.

Images