Packaging detection device based on AI technology

Abstract

The utility model discloses an AI technology packaging detection device which comprises a vertical cabinet, an assembly line is arranged in a channel in the side face of the vertical cabinet, a through groove is formed in the assembly line, a left limiting block and a right limiting block are symmetrically arranged in the channel, and the outer sides of the left limiting block and the right limiting block are connected with first telescopic rods. A front-back limiting block is arranged on the side face of the left-right limiting block, a second telescopic rod is connected to the top of the front-back limiting block, a first camera and a second camera are arranged on the upper side and the lower side of the assembly line correspondingly, a waste removing mechanism is arranged along one side of the assembly line, and a visual controller and a PLC are further arranged in the vertical cabinet. According to the utility model, the structure design is scientific and reasonable, objects entering the vertical cabinet can be automatically detected, the to-be-detected objects are collected through the cooperation of the two cameras, the to-be-detected objects are judged through the visual controller, the purpose of rapidly detecting the product sealing qualification is achieved, and the detection efficiency is improved.

Description

Technical Field

[0001] The utility model relates to the technical field of packaging detection equipment, in particular to a packaging detection device using AI technology. Background Art

[0002] In the packaging process of daily necessities, such as diapers, sanitary napkins and other commonly used packaging bag products, the packaging process includes several key steps, including bagging, opening, filling, heat sealing, quality inspection and other links. Especially in the final quality inspection stage, since the packaging bags that are not completely closed or partially sealed are almost the same as qualified products in appearance, this leads to a problem: it is difficult for standard inspection processes to distinguish them. At present, this process often relies on a large number of manual resources to perform, and staff need to evaluate the sealing quality of each packaging bag through visual inspection. They even need to actually check each bag with their fingers to determine whether the seal is complete. This method is not only inefficient, but also highly dependent on personal judgment, which brings considerable subjectivity, further leading to quality assurance problems.

[0003] In this case, the application of automation and machine vision systems has become a potential solution. By using high-precision industrial cameras and advanced image analysis software, sealing problems can be automatically identified, reducing reliance on manual inspections, thereby improving production efficiency and consistency of product quality. To this end, we propose a packaging inspection device based on AI technology. Utility Model Content

[0004] The purpose of the utility model is to provide a packaging detection device based on AI technology to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned purpose, the utility model provides the following technical solutions: a packaging detection device using AI technology, comprising a cabinet, an assembly line is arranged in a channel on the side of the cabinet, a through slot is opened on the assembly line, left and right limit blocks are symmetrically arranged in the channel, the outer sides of the left and right limit blocks are connected to a first telescopic rod, front and rear limit blocks are arranged on the sides of the left and right limit blocks, and the tops of the front and rear limit blocks are connected to a second telescopic rod, a first camera and a second camera are respectively arranged on the upper and lower sides of the assembly line, a waste rejection mechanism is arranged along one side of the assembly line, and a visual controller and a PLC are also arranged inside the cabinet.

[0006] In the above solution, a side door is installed on the side of the cabinet, and a support block is arranged on the bottom of the cabinet.

[0007] In the above solution, an alarm is installed on the top of the cabinet.

[0008] In the above solution, a display is installed on the side of the cabinet via a column.

[0009] In the above solution, infrared light sources are respectively arranged on the sides of the first camera and the second camera.

[0010] In the above solution, the visual controller and the PLC are connected to a power source via a wire.

[0011] In the above solution, the waste rejection mechanism is an air blowing waste rejection mechanism, and a waste collection bin is provided on the side of the waste rejection mechanism.

[0012] Compared with the prior art, the beneficial effects of the utility model are as follows: the packaging inspection device of this AI technology has a simple and reasonable structural design and strong practicality. By arranging an assembly line, left and right limit blocks, front and rear limit blocks, a first camera, a second camera, a waste rejection mechanism, a visual controller and a PLC in the cabinet, the objects entering the cabinet can be automatically inspected. The objects to be inspected are collected through the cooperation of the two cameras and judged through the visual controller, so as to achieve the purpose of quickly detecting the sealing qualification of the product and improve the inspection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the structure of the utility model.

[0014] Figure 2 It is a schematic diagram of the internal structure of the utility model.

[0015] Figure 3 This is a schematic diagram of the material limiting structure of the utility model.

[0016] In the figure: 1, cabinet 11, passage 12, assembly line 13, through slot 14, side door 15, support block 16, alarm 17, column 18, display 19, left and right limit blocks 2, first telescopic rod 21, front and rear limit blocks 22, second telescopic rod 23, first camera 24, second camera 25, infrared light source 26, waste rejection mechanism 27, visual controller 28, PLC. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the utility model to clearly and completely describe the technical solutions in the embodiments of the utility model. Obviously, the described embodiments are only part of the embodiments of the utility model, not all of the embodiments. Based on the embodiments in the utility model, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the utility model.

[0018] See also Figure 1-3The utility model provides a technical solution: a packaging detection device using AI technology, comprising a cabinet 1, an assembly line 12 is arranged in a channel 11 on the side of the cabinet 1, a through slot 13 is opened on the assembly line 12, and the design of the through slot 13 enables a second camera 24 to effectively observe the object from below, left and right limit blocks 19 are symmetrically arranged in the channel 11, the outer sides of the left and right limit blocks 19 are connected to a first telescopic rod 2, front and rear limit blocks 21 are arranged on the sides of the left and right limit blocks 19, and a second telescopic rod 22 is connected to the top of the front and rear limit blocks 21, a first camera 23 and a second camera 24 are respectively arranged on the upper and lower sides of the assembly line 12, a waste rejection mechanism 26 is arranged along one side of the assembly line 12, and a visual controller 27 and a PLC 28 are also arranged inside the cabinet 1.

[0019] In the above solution, a side door 14 is installed on the side of the cabinet 1 , and a support block 15 is provided on the bottom of the cabinet 1 .

[0020] In the above solution, an alarm 16 is installed on the top of the cabinet 1. When the visual controller 27 detects that the sealing of the object is unqualified, the alarm 16 is used to give an alarm prompt.

[0021] In the above solution, a display 18 is installed on the side of the cabinet 1 through a column 17.

[0022] In the above solution, infrared light sources 25 are respectively disposed on the sides of the first camera 23 and the second camera 24. The infrared light sources 25 are used to supplement the light source to ensure the image quality of the first camera 23 and the second camera 24.

[0023] In the above solution, the visual controller 27 and PLC 28 are connected to a power source via a wire.

[0024] In the above solution, the waste rejection mechanism 26 is an air blowing waste rejection mechanism, and a waste collection bin is provided on the side of the waste rejection mechanism 26. The waste rejection mechanism 26 is an existing mechanism, which rejects waste products by generating high-speed high-pressure gas.

[0025] Working principle:

[0026] The packaging inspection device based on AI technology has the following specific operations:

[0027] 1. The visual inspection system is installed in a cabinet 1;

[0028] 2. There are 40mmx40mm spaces on the left and right sides of cabinet 1 to connect the front and rear assembly lines;

[0029] 3. The cabinet 1 has double doors at the front and rear to facilitate debugging and maintenance, and the display 18 is hung on the outer column 17;

[0030] 4. When switching products, the defect-free object 1 passes through the left and right limit blocks 19 from the left assembly line to correct the posture of the object and enter the visual inspection device from the front. Then, under the action of the second telescopic rod 22, the front and rear limit blocks 21 block the corrected object 1 from top to bottom, and send a signal to the first camera 23 and the second camera 24 to take pictures. Then, the front and rear limit blocks 21 move from bottom to top to release the object 1 and continue to move forward along the track. The first camera 23 and the second camera 24 take pictures of the object 1 to obtain the upper image 1 to be learned and the lower image 2 to be learned;

[0031] 5. The upper image 1 and the lower image 2 are respectively transmitted to the visual controller 27 (X86 architecture industrial computer) through the network cable for unsupervised learning to obtain model 1 and model 2, such as the TadGAN algorithm; the process from 4 to 5 is called the self-learning mode;

[0032] 6. Then enter the detection mode and start collecting the object to be detected from step 4 to obtain the upper image 1 and the lower image 2 to be detected;

[0033] 7. The upper image 1 and the lower image 2 to be inspected are transmitted to the visual controller (X86 architecture industrial computer) to read the model 1 and the model 2 respectively to identify whether the object is OK or NG, that is, to confirm whether the outer packaging seal on the surface of the object is qualified;

[0034] 8. Send the recognition result signal to PLC28 to control whether the waste rejection mechanism 26 is started. When the OK signal is received, the waste rejection mechanism 26 is not started. When the NG signal is received, the waste rejection mechanism 26 is started. Under the action of strong airflow, the unqualified objects 1 are removed into the waste collection bin.

[0035] 9. Repeat steps 6 to 9 until all products of this model have been produced and inspected. This method has several significant advantages: a. The unsupervised GAN method has the advantage of rapid learning when changing products, while solving the problem of poor detection effect of traditional detection algorithms; b. It can solve the problem that supervised learning methods have a long training time and cannot adapt to rapid product changes, especially when the product changes more than 5 times a day; c. The visual inspection device uses upper and lower linear array cameras with infrared light sources to perform all-round inspection of the patterns on the upper and lower sides of the packaging bag and the damage inside the bag, with the advantages of low cost and good effect.

[0036] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A packaging inspection device using AI technology, comprising a cabinet (1), characterized in that: An assembly line (12) is arranged in a channel (11) on the side of the cabinet (1), a through slot (13) is provided on the assembly line (12), left and right limit blocks (19) are symmetrically arranged in the channel (11), the outer sides of the left and right limit blocks (19) are connected to a first telescopic rod (2), front and rear limit blocks (21) are arranged on the sides of the left and right limit blocks (19), and the tops of the front and rear limit blocks (21) are connected to a second telescopic rod (22), the upper and lower sides of the assembly line (12) are respectively provided with a first camera (23) and a second camera (24), a waste rejection mechanism (26) is arranged along one side of the assembly line (12), and a visual controller (27) and a PLC (28) are also arranged inside the cabinet (1).

2. The packaging detection device using AI technology according to claim 1, characterized in that: A side door (14) is installed on the side of the cabinet (1), and a support block (15) is arranged at the bottom of the cabinet (1).

3. The packaging detection device using AI technology according to claim 1, characterized in that: An alarm (16) is installed on the top of the cabinet (1).

4. The packaging detection device using AI technology according to claim 1, characterized in that: A display (18) is installed on the side of the cabinet (1) via a column (17).

5. The packaging detection device using AI technology according to claim 1, characterized in that: Infrared light sources (25) are respectively arranged on the sides of the first camera (23) and the second camera (24).

6. The packaging detection device using AI technology according to claim 1, characterized in that: The visual controller (27) and the PLC (28) are connected to a power source via a wire.

7. The packaging detection device using AI technology according to claim 1, characterized in that: The waste rejection mechanism (26) is an air blowing waste rejection mechanism, and a waste collection bin is provided on the side of the waste rejection mechanism (26).

Images