Paper box non-defective product detection mechanism

Abstract

The utility model discloses a paper box non-defective product detection mechanism which comprises at least one detection assembly, each detection assembly comprises four camera frames arranged at four corners, a connecting frame movably connected with the corresponding camera frame and a mounting rod group for fixedly mounting the connecting frame, the two camera frames and the corresponding connecting frames form one group and are mounted on the mounting rod group on one side, the other two camera frames and the corresponding connecting frames form the other group and are mounted on the mounting rod group on the other opposite side, the two mounting rod groups can move towards each other or away from each other, and a visual camera is mounted on each camera frame. The paper box edge folding detection device is reasonable in structural design, the quality of the outer edge folding and the inner edge folding of a paper box can be well detected by arranging at least one detection assembly, and the product quality can be better guaranteed.

Description

Technical Field

[0001] The utility model relates to the technical field of packaging equipment, and more specifically, to a detection mechanism for good-quality paper boxes. Background Art

[0002] In the prior art, for paper boxes (such as lidless boxes, mobile phone boxes, etc.) that have been packaged and have openings, it is necessary to detect the quality of the finished products. For example, the detection of the outer folding edge angle of the paper box, the detection of the outer folding edge wrinkles, and when the paper box has an inward folding edge, it is also necessary to detect the inner folding edge angle, the inner folding edge covering quality, and the inner folding edge wrinkles, etc. Content of the Utility Model

[0003] The utility model overcomes the deficiencies of the prior art, has a reasonable structural design, and by setting at least one set of detection components, it can well detect the quality of the outer folding edge and the inner folding edge of the paper box, and can better ensure the product quality.

[0004] To achieve the above object, the utility model provides the following technical solutions:

[0005] A detection mechanism for good-quality paper boxes includes at least one detection component. Each detection component includes four camera mounts arranged at the four corners, a connecting frame movably connected to the corresponding camera mount, and a set of mounting rods for fixedly installing the connecting frame. Two camera mounts and the corresponding connecting frame are installed on a set of mounting rods on one side, and the other two camera mounts and the corresponding connecting frame are installed on another set of mounting rods on the opposite side. The two sets of mounting rods can move towards or away from each other, and a vision camera is installed on each camera mount.

[0006] Preferably, a receiving groove is provided at one end of each camera mount close to the connecting frame, and the connecting frame is connected with a universal ball head that is adapted to the receiving groove and is located in the receiving groove.

[0007] Preferably, an illuminating light source is provided at the end of the camera mount far from the receiving groove.

[0008] Preferably, a guiding rod is connected between the two sets of mounting rods, and a driving component for driving the mounting rods to move along the guiding rod is provided on one side of at least one set of mounting rods.

[0009] Preferably, two detection components are provided, and the two detection components are arranged in an upper and lower layer. The detection component located in the upper layer is used to photograph the inner side of the paper box, and the detection component located in the lower layer is used to photograph the outer side of the paper box.

[0010] Preferably, the four mounting rods of the two detection components are located on the same straight line, and the four mounting rods are connected by a connecting rod set.

[0011] Advantages of the present utility model: The structure of the present utility model is reasonably designed. By setting at least one set of detection components, the quality of the outer folding edges and inner folding edges of the paper box can be well detected, which can better ensure the product quality. Description of the Drawings

[0012] Figure 1 It is a schematic structural diagram of a specific embodiment of the present utility model when there are two detection components;

[0013] Figure 2 It is a schematic structural diagram of a specific embodiment of the present utility model when there is one detection component;

[0014] Figure 3 It is a schematic connection structure diagram of a camera holder and a connecting frame in a specific embodiment of the present utility model;

[0015] Figure 4 It is a schematic diagram of the use state of a specific embodiment of the present utility model when there is one detection component.

[0016] In the figure: 100, conveyor belt; 101, paper box; 1, detection component; 2, camera holder; 21, vision camera; 22, accommodation groove; 23, illuminating light source; 3, connecting frame; 31, universal ball head; 4, mounting rod group; 5, guide rod; 6, driving component; 7, connecting rod group. Detailed Embodiments

[0017] Next, the technical solutions in the embodiments of the present utility model will be clearly and completely described in conjunction with the drawings in the embodiments of the present utility model. Obviously, the described embodiments are only a part of the embodiments of the present utility model, rather than all the embodiments. Based on the embodiments of the present utility model, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present utility model.

[0018] Such as Figures 1-4As shown in the figure, a good product detection mechanism for paper boxes includes at least one detection component 1. Each detection component 1 includes four camera mounts 2 arranged at the four corners, a connecting frame 3 movably connected to the corresponding camera mount 2, and a mounting rod group 4 for fixedly installing the connecting frame 3. Two camera mounts 2 and the corresponding connecting frame 3 are installed on one side of the mounting rod group 4 as a group, and the other two camera mounts 2 and the corresponding connecting frame 3 are installed on the mounting rod group 4 on the opposite side as another group. The two mounting rod groups 4 can move towards or away from each other. A vision camera 21 is installed on each camera mount 2; a receiving groove 22 is provided at one end of each camera mount 2 close to the connecting frame 3, and the connecting frame 3 is connected with a universal ball head 31 that is adapted to the receiving groove 22 and is located in the receiving groove 22; a lighting source 23 is provided at the end of the camera mount 2 far from the receiving groove 22; a guiding rod 5 is connected between the two mounting rod groups 4, and a driving component 6 for driving one of the mounting rod groups 4 to move along the guiding rod 5 is provided on one side of at least one mounting rod group 4; there are two detection components 1, and the two detection components 1 are arranged in an upper and lower layer. The detection component 1 located in the upper layer is used to photograph the inner side of the paper box 101, and the detection component 1 located in the lower layer is used to photograph the outer side of the paper box 101; the four mounting rod groups 4 of the two detection components 1 are on the same straight line, and the four mounting rod groups 4 are connected by a connecting rod group 7.

[0019] By adopting the above technical solution, in the present utility model, during the actual production process, the detection component 1 is arranged above the conveyor belt 100 and is used to detect the paper box 101. When there is one detection component 1, it can realize photographing and detecting the four outer corners of the paper box 101, and the four vision cameras 21 can be aligned with the four outer corners of the paper box 101 at a 45-degree angle. When there are two detection components 1, the four vision cameras of the lower detection component 1 are aligned with the four outer corners of the paper box 101, and the four cameras of the upper detection component 1 are aligned with the four inner corners of the paper box 101 to photograph and detect the quality of the inner corner folding edges and the outer corner folding edges of the paper box 101.

[0020] Among them, by providing the driving component 6 and the guiding rod 5, the driving component 6 can be set as a common linear driver such as a linear motor or a cylinder, which can realize the relative movement of the two mounting rod groups 4 towards or away from each other, that is, the translational adjustment of the camera mount 2 can be realized to meet the photographing and detecting requirements for paper boxes 101 of different sizes.

[0021] By providing the universal ball head 31 and the receiving groove 22, the universal movement adjustment of the camera mount 2 relative to the connecting frame 3 can be realized, so that the vision camera 21 can have an accurate angle of alignment with the paper box 101. And by providing the lighting source 23 on the camera mount 2, specifically an LED lamp, clearer photographing and imaging of the vision camera 21 can be realized.

[0022] And it is worth mentioning that the vision camera 21 of the present utility model can change the iterative learning that takes a long time for traditional defective product detection by detecting good products. The specific detection steps can be as follows: First, manually check and confirm 10 - 50 qualified products, then select the filtered picture folder from the software and directly conduct deep learning training without annotation. After about 5 minutes of training, load the model to complete the creation of a new product. The detection processes of all vision cameras 21 are the same. Use unsupervised deep learning tools to automatically determine whether the detection target is qualified based on the trained qualified pictures. For products that can be manually confirmed as qualified products, just put the product pictures into the training folder for additional training. As for the specific process and principle from shooting to output: The PLC controls multiple groups of vision cameras 21 to shoot according to the trigger signal of the photoelectric sensor, and then sends the current product flag bit to the detection software through communication; after the detection software collects and processes the images, it integrates the results according to the current processed product flag and then sends them to the PLC, and the PLC controls whether to reject the product.

[0023] The above description of the specific principle is only one detection mode that the present utility model can adopt. The main protection point of the present utility model still lies in the structure itself.

[0024] The above is only the preferred specific implementation manner of the present utility model, but the protection scope of the present utility model is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present utility model, according to the technical solution of the present utility model and its inventive concept, makes equivalent substitutions or changes, should be covered within the protection scope of the present utility model.

Claims

1. A carton good product detection mechanism, characterized in that: The invention comprises at least one detection assembly (1), each detection assembly (1) comprising four camera frames (2) arranged at four corners, a connecting frame (3) movably connected to the corresponding camera frames (2), and a mounting rod group (4) for fixing the connecting frame (3); two camera frames (2) and the corresponding connecting frames (3) are mounted as a group on the mounting rod group (4) on one side, and the other two camera frames (2) and the corresponding connecting frames (3) are mounted as another group on the mounting rod group (4) on the opposite side; the two mounting rod groups (4) can move towards or away from each other, and each camera frame (2) is mounted with a visual camera (21).

2. A paper box good product detection mechanism according to claim 1, characterized in that: Each camera frame (2) is provided with a receiving groove (22) at one end close to the connecting frame (3), and the connecting frame (3) is connected to a universal ball head (31) adapted to the receiving groove (22) and located in the receiving groove (22).

3. A carton good product detection mechanism according to claim 1 or 2, characterized in that: An illuminating light source (23) is provided on one end of the camera frame (2) away from the accommodating groove (22).

4. A paper box good product detection mechanism according to claim 3, characterized in that: A guide rod (5) is connected between the two installation rod groups (4), and a driving component (6) for driving the installation rod group (4) to move along the guide rod (5) is provided on one side of at least one installation rod group (4).

5. A paper box good product detection mechanism according to claim 4, characterized in that: Two detection components (1) are provided, and the two detection components (1) are arranged in upper and lower layers. The detection component (1) located in the upper layer is used to photograph the inner side of the paper box (101), and the detection component (1) located in the lower layer is used to photograph the outer side of the paper box (101).

6. A paper box good product detection mechanism according to claim 5, characterized in that: The four installation rod groups (4) of the two detection components (1) are located on the same straight line, and the four installation rod groups (4) are connected by a connecting rod group (7).

Images