Package printed matter detection device based on deep learning

Abstract

The utility model relates to the technical field of deep learning, in particular to a packaging printed matter detection device based on deep learning, which comprises a base, a frame body fixedly mounted at the top of the base, a CCD (Charge Coupled Device) camera fixedly mounted at the bottom of the frame body, and a rotating component used for positioning and rotationally detecting a packaging printed product to be detected, the rotating assembly comprises two fixing plates, and a rotating shaft is rotationally arranged on one side of each fixing plate; by opening the two second electric push rods, the driving ends of the two second electric push rods can extend to drive the two positioning pieces to get close to each other and clamp a packaged product, then the first electric push rod is opened in a matched mode, the supporting plate can slide downwards in the sliding opening, and then the third electric push rod is opened; the rack moves in the fixing cavity and drives the gear, the rotating shaft, the positioning piece and the packaged product to rotate, so that the CCD camera can scan and detect the periphery of the packaged product, the detection range is widened, and the detection effect is improved.

Description

Technical Field

[0001] The utility model relates to the technical field of deep learning, and in particular to a packaging printed matter detection device based on deep learning. Background Art

[0002] After the packaging printed matter is printed, it is necessary to inspect the printing quality. The existing inspection devices on the market usually use the most common camera inspection structure, which has low recognition accuracy and low recognition power for the printed matter. Therefore, there are now high-definition inspection devices for packaging printed matter.

[0003] After checking, the public announcement number: CN220289442U discloses a packaging printed product inspection device based on deep learning. This technology discloses that "through a high-resolution camera and image processing software, the system can quickly capture and analyze printed images. With the addition of artificial intelligence and deep learning technology, the device can more accurately identify and classify printing defects, and can even identify complex and tiny defects."

[0004] However, the above-mentioned existing deep learning-based packaging printed matter detection device still has some shortcomings when in use: although the deep learning-based packaging printed matter detection device can perform high-definition detection of packaged products, its detection range is small. Since the camera is set on the top of the product, it is not convenient to detect other sides of the product. Although the camera on the deep learning-based packaging printed matter detection device is movable, it is still unable to scan clearly because the camera is always facing downward.

[0005] In order to solve the above problems, this application proposes a packaging printed matter detection device based on deep learning. Utility Model Content

[0006] In order to solve the problems raised in the above background technology, the utility model provides a packaging printed product detection device based on deep learning, which can conveniently detect the four sides of the packaging product and conveniently adjust the conveying direction of the packaging product.

[0007] To achieve the above-mentioned purpose, the utility model provides the following technical solutions: a packaging printed product detection device based on deep learning, comprising a base, a frame body is fixedly installed on the top of the base, a CCD camera is fixedly installed on the bottom of the frame, and a rotating component is also included for positioning and rotating the packaging printed product to be detected for detection;

[0008] The rotating assembly includes two fixed plates, a rotating shaft is rotatably provided on one side of the fixed plate, a second electric push rod is fixedly installed on the other end of the rotating shaft, a positioning piece is provided on the driving end of the second electric push rod, one of the rotating shafts rotates through the fixed plate and extends to the inside of the fixed plate and a gear is fixedly sleeved thereon, a third electric push rod is fixedly installed on one side of one of the fixed plates, a rack is provided on the driving end of the third electric push rod, the rack is meshed with the gear, a first electric push rod is provided in the base, a sliding opening is opened on the top of the base, a support plate is slidably connected in the sliding opening, and the support plate is connected to the driving end of the first electric push rod.

[0009] As a preferred deep learning-based packaging printed product detection device of the utility model, a fixed cavity is opened inside one of the fixed plates, one end of one of the rotating shafts is rotatably connected in the fixed cavity, and the rack is slidably connected in the fixed cavity.

[0010] As a preferred embodiment of the utility model as a packaging printed product detection device based on deep learning, a conveying assembly is arranged on the top of the base, and the conveying assembly includes a belt, a motor and two rotating rollers, the two rotating rollers are symmetrically rotatably arranged in the base, the belt tensioning sleeve is connected to the two rotating rollers, the upper part of the belt is located outside the base, the motor is arranged on the front side of the base, and the driving shaft of the motor is transmission-connected to one end of one of the rotating rollers, which can facilitate and automatically convey the packaged products, thereby reducing the workload of personnel.

[0011] As a preferred embodiment of the deep learning-based packaging printed product detection device of the utility model, a sliding cavity and a limiting opening are respectively opened inside and on the top of the base, the first electric push rod is fixedly installed in the sliding cavity, and the rotating roller is rotatably connected in the limiting opening.

[0012] As a preferred embodiment of the deep learning-based packaged printed product detection device of the utility model, it also includes a guide component for guiding the packaged products conveyed on the belt, the guide component includes two limit bars, the two limit bars are arranged on the top of the base and on both sides of the belt, the top of the limit bar is rotatably connected to the transmission shaft, the top of the transmission shaft is fixedly installed with a guide member, a motor is embedded in the limit bar, and the driving shaft of the motor is transmission-connected to one end of the transmission shaft. When the packaged product is conveyed on the belt, the two motors are turned on at the same time, so that the two transmission shafts drive the guide members to rotate, so that the two guide members form an eight-shaped shape, so that the packaged product is conveyed to the center position of the belt along the space between the two guide members and straightened, which can help the two positioning members to clamp the packaged product.

[0013] As a preferred deep learning-based packaging printed product detection device of the utility model, a display controller is arranged on the front side of the base, and the display controller is electrically connected to the CCD camera, the first electric push rod, the second electric push rod, the third electric push rod and the motor, so that the detection results can be conveniently reflected on the display controller, and personnel can also conveniently control the switch and operating status of the CCD camera, the first electric push rod, the second electric push rod, the third electric push rod and the motor.

[0014] Compared with the prior art, the beneficial effects of the utility model are:

[0015] 1. On this basis, a rotating component is added. By turning on the two second electric push rods, the driving ends of the two second electric push rods can be extended to drive the two positioning pieces to approach and clamp the packaged product. Then, the first electric push rod is turned on to make the support plate slide downward in the sliding mouth. Then, the third electric push rod is turned on to make the rack move in the fixed cavity and drive the gear, the rotating shaft, the positioning piece and the packaged product to rotate, so that the CCD camera can scan and detect all around the packaged product, thereby increasing the detection range and thus improving the detection effect.

[0016] 2. At the same time, a guide component is added on this basis. By turning on the two motors, the two transmission shafts can drive the guide parts to rotate, so that the two guide parts form an eight-shaped shape, so that the packaged products are transported to the center position of the belt along the space between the two guide parts and aligned, which can help the two positioning parts to clamp the packaged products. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0018] Figure 1 It is a schematic diagram of the overall structure of the utility model;

[0019] Figure 2 It is a three-dimensional structural schematic diagram of the interior of the base of the utility model in a half-section view;

[0020] Figure 3 It is a three-dimensional structural schematic diagram of the interior of the fixing plate of the utility model in a half-section view;

[0021] Figure 4 It is a three-dimensional structural schematic diagram of the interior of the limiting strip of the utility model in a half-section view.

[0022] In the figure:

[0023] 1. Base; 2. Frame; 3. CCD camera; 4. First electric push rod; 5. Support plate; 6. Fixed plate; 7. Rotating shaft; 8. Second electric push rod; 9. Positioning member; 10. Gear; 11. Third electric push rod; 12. Rack; 13. Fixed cavity; 14. Limiting strip; 15. Transmission shaft; 16. Guide member; 17. Motor; 18. Rotating roller; 19. Belt; 20. Motor; 21. Sliding cavity; 22. Sliding mouth; 23. Limiting mouth. DETAILED DESCRIPTION

[0024] 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.

[0025] Example 1

[0026] like Figure 1-Figure 4 As shown;

[0027] A packaging printed product detection device based on deep learning comprises a base 1, a frame 2 is fixedly mounted on the top of the base 1, and a CCD camera 3 is fixedly mounted on the bottom of the frame 2.

[0028] In this implementation scheme: After checking the public (announcement) number: CN220289442U discloses a packaging printed product inspection device based on deep learning. This technology discloses that "through a high-resolution camera and image processing software, the system can quickly capture and analyze printed images. By adding artificial intelligence and deep learning technology, the device can more accurately identify and classify printing defects, and can even identify complex and tiny defects."

[0029] The existing deep learning-based packaging printed matter detection device still has some shortcomings when in use: although the deep learning-based packaging printed matter detection device can perform high-definition detection of packaged products, its detection range is small. Since the camera is set on the top of the product, it is not convenient to detect other sides of the product. Although the camera on the deep learning-based packaging printed matter detection device is movable, since the camera is always facing downward, it still cannot scan clearly. Combined with actual use, this problem is obviously a real problem that is difficult to solve. In view of this, in order to solve this technical problem, a rotating component and a guide component are added to the present application document.

[0030] According to the above content, in order to facilitate the inspection of the printing on all sides of the packaging product, a rotating component is also included to position and rotate the packaging printing product to be inspected;

[0031] The rotating assembly includes two fixed plates 6, a rotating shaft 7 is rotatably provided on one side of the fixed plate 6, a second electric push rod 8 is fixedly installed on the other end of the rotating shaft 7, a positioning member 9 is provided on the driving end of the second electric push rod 8, one of the rotating shafts 7 rotates through the fixed plate 6 and extends to the inside of the fixed plate 6 and a gear 10 is fixedly sleeved thereon, a third electric push rod 11 is fixedly installed on one side of one of the fixed plates 6, a rack 12 is provided on the driving end of the third electric push rod 11, the rack 12 is meshed and connected with the gear 10, a first electric push rod 4 is provided in the base 1, a sliding opening 22 is provided on the top of the base 1, a support plate 5 is slidably connected in the sliding opening 22, and the support plate 5 is connected to the driving end of the first electric push rod 4, a fixed cavity 13 is provided in the interior of one of the fixed plates 6, one end of one of the rotating shafts 7 is rotatably connected in the fixed cavity 13, and the rack 12 is slidably connected in the fixed cavity 13;

[0032] A conveying assembly is arranged on the top of the base 1, and the conveying assembly includes a belt 19, a motor 20 and two rotating rollers 18. The two rotating rollers 18 are symmetrically rotatably arranged in the base 1, and the belt 19 is tensioned and sleeved on the two rotating rollers 18. The upper half of the belt 19 is located outside the base 1. The motor 20 is arranged on the front side of the base 1, and the driving shaft of the motor 20 is transmission-connected to one end of one of the rotating rollers 18. A sliding cavity 21 and a limiting opening 23 are respectively opened inside and on the top of the base 1. The first electric push rod 4 is fixedly installed in the sliding cavity 21, and the rotating roller 18 is rotatably connected in the limiting opening 23.

[0033] In this embodiment: the packaged product to be inspected is placed on the top of the belt 19, and then the motor 20 is turned on to rotate the rotating roller 18, so that the belt 19 moves and drives the packaged product to move toward the frame 2 until the packaged product slides to the top of the support plate 5, and then the two second electric push rods 8 are turned on, so that the driving ends of the two second electric push rods 8 are extended to drive the two positioning members 9 to approach and clamp the packaged product, and then the first electric push rod 4 is turned on, and the driving end of the first electric push rod 4 is shortened, so that the support plate 5 slides downward in the sliding mouth 22, and then the third electric push rod 11 is turned on, and the driving end of the third electric push rod 11 is extended to drive the rack 12 to move in the fixed cavity 13, so that the rack 12 drives the gear 10, the rotating shaft 7, the positioning member 9 and the packaged product to rotate, so that the CCD camera 3 can scan and detect all around the packaged product, thereby increasing the detection range and thus improving the detection effect. When the rack 12 is against the end of the fixed cavity 13, the packaged product just rotates one circle and returns to its position.

[0034] In an optional embodiment: it also includes a guide component for guiding the packages conveyed on the belt 19, the guide component includes two limit bars 14, the two limit bars 14 are both arranged on the top of the base 1 and located on both sides of the belt 19, the top of the limit bar 14 is rotatably connected to the transmission shaft 15, the top of the transmission shaft 15 is fixedly installed with a guide member 16, the limit bar 14 is embedded with a motor 17, and the drive shaft of the motor 17 is transmission-connected to one end of the transmission shaft 15.

[0035] In this embodiment: on the basis of the above implementation, when the packaged product is conveyed on the belt 19, the two motors 17 are turned on at the same time, so that the two transmission shafts 15 drive the guide members 16 to rotate, so that the two guide members 16 form an eight-shaped shape, so that the packaged product is conveyed along the space between the two guide members 16 to the center position of the belt 19 and aligned, which can help the two positioning members 9 to clamp the packaged product.

[0036] In an optional embodiment: a display controller is disposed on the front side of the base 1 , and the display controller is electrically connected to the CCD camera 3 , the first electric push rod 4 , the second electric push rod 8 , the third electric push rod 11 and the motor 17 .

[0037] In this embodiment, the detection result can be conveniently reflected on the display controller, and the personnel can also conveniently control the switch and operation status of the CCD camera 3, the first electric push rod 4, the second electric push rod 8, the third electric push rod 11 and the motor 17.

[0038] The working principle and use process of the utility model are as follows: first, the electrical appliances involved in the packaging printed product detection device based on deep learning are connected to an external power supply, and then the packaged product to be detected is placed on the top of the belt 19, and then the motor 20 is turned on to rotate the rotating roller 18, so that the belt 19 moves and drives the packaged product to move toward the frame 2 until the packaged product slides to the top of the support plate 5, and then the two second electric push rods 8 are turned on to extend the driving ends of the two second electric push rods 8 to drive the two positioning members 9 to approach and clamp the packaged product, and then the first electric push rod 4 is turned on to shorten the driving end of the first electric push rod 4 so that the support plate 5 slides downward in the sliding opening 22, and then the third electric push rod 11 is turned on to extend the driving end of the third electric push rod 11 The elongated end drives the rack 12 to move in the fixed cavity 13, so that the rack 12 drives the gear 10, the rotating shaft 7, the positioning member 9 and the packaged product to rotate, so that the CCD camera 3 can scan and detect all around the packaged product, thereby increasing the detection range and thus improving the detection effect. When the rack 12 abuts against the end of the fixed cavity 13, the packaged product just rotates one circle and returns to its position. In addition, when the above-mentioned packaged product is conveyed on the belt 19, the two motors 17 are turned on at the same time, so that the two transmission shafts 15 drive the guide members 16 to rotate, so that the two guide members 16 form an eight-shaped shape, so that the packaged product is conveyed to the center position of the belt 19 along the space between the two guide members 16 and is aligned, which can help the two positioning members 9 to clamp the packaged product.

[0039] Finally, it should be noted that the above description is only a preferred embodiment of the utility model and is not intended to limit the utility model. Although the utility model is described in detail with reference to the above embodiments, those skilled in the art can still modify the technical solutions recorded in the above embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the utility model shall be included in the protection scope of the utility model.

Claims

1. A packaging printed product detection device based on deep learning, comprising a base (1), a frame (2) fixedly mounted on the top of the base (1), and a CCD camera (3) fixedly mounted on the bottom of the frame (2), characterized in that: It also includes a rotating component for positioning and rotating the packaging and printing product to be inspected; The rotating assembly comprises two fixed plates (6), one side of the fixed plate (6) is rotatably provided with a rotating shaft (7), the other end of the rotating shaft (7) is fixedly installed with a second electric push rod (8), the driving end of the second electric push rod (8) is provided with a positioning member (9), one of the rotating shafts (7) rotates through the fixed plate (6) and extends into the interior of the fixed plate (6) and a gear (10) is fixedly sleeved thereon, one side of the fixed plate (6) is fixedly installed with a third electric push rod (11), the driving end of the third electric push rod (11) is provided with a rack (12), the rack (12) is meshedly connected with the gear (10), a first electric push rod (4) is arranged in the base (1), a sliding opening (22) is opened at the top of the base (1), a support plate (5) is slidably connected in the sliding opening (22), and the support plate (5) is connected to the driving end of the first electric push rod (4).

2. The packaging printed matter detection device based on deep learning according to claim 1, characterized in that: A fixed cavity (13) is provided inside one of the fixed plates (6), one end of one of the rotating shafts (7) is rotatably connected in the fixed cavity (13), and the rack (12) is slidably connected in the fixed cavity (13).

3. The packaging printed matter detection device based on deep learning according to claim 1, characterized in that: A conveying assembly is arranged on the top of the base (1), and the conveying assembly comprises a belt (19), a motor (20) and two rotating rollers (18). The two rotating rollers (18) are symmetrically arranged in the base (1) for rotation. The belt (19) is tensioned and sleeved on the two rotating rollers (18). The upper part of the belt (19) is located outside the base (1). The motor (20) is arranged on the front side of the base (1), and the driving shaft of the motor (20) is drivingly connected to one end of one of the rotating rollers (18).

4. The packaging printed matter detection device based on deep learning according to claim 3 is characterized in that: The base (1) is provided with a sliding cavity (21) and a limiting opening (23) inside and on the top thereof, respectively; the first electric push rod (4) is fixedly installed in the sliding cavity (21); and the rotating roller (18) is rotatably connected in the limiting opening (23).

5. The packaging printed matter detection device based on deep learning according to claim 4, characterized in that: The invention also comprises a guide assembly for guiding the packaged products conveyed on the belt (19), wherein the guide assembly comprises two limit bars (14), both of which are arranged on the top of the base (1) and located on both sides of the belt (19), the top of the limit bar (14) is rotatably connected to a transmission shaft (15), the top of the transmission shaft (15) is fixedly mounted with a guide member (16), a motor (17) is embedded in the limit bar (14), and the drive shaft of the motor (17) is transmission-connected to one end of the transmission shaft (15).

6. The packaging printed matter detection device based on deep learning according to claim 5, characterized in that: A display controller is arranged on the front side of the base (1), and the display controller is electrically connected to the CCD camera (3), the first electric push rod (4), the second electric push rod (8), the third electric push rod (11) and the motor (17).

Images