Defective product removing structure with AI visual scanning

Abstract

The utility model provides a kind of defective product rejection structure with AI vision scanning, including conveying belt, conveying power structure, AI vision equipment, rejection structure, conduction path, send back conveyor and send back power structure, wherein: conveying power structure is installed in one end of conveying belt, and AI vision equipment is fixedly installed in the outside of conveying belt feeding end, and the rejection structure is fixedly installed in the side of conveying belt discharge end;The conduction path is fixedly installed in the other side of conveying belt discharge end, and send back conveyor is located below the side of conveying belt, and the send back power structure is installed in one end of send back conveyor;The setting of AI vision equipment and send back conveyor of the utility model can effectively detect defective product, and there is no misjudgment when detecting.

Description

Technical Field

[0001] This utility model relates to the field of defective product rejection technology, and in particular to a defective product rejection structure with AI visual scanning. Background Technology

[0002] In industrial production, especially on product manufacturing and packaging lines, quality inspection and rejection of defective products are crucial steps in ensuring the quality of the final product. Traditional manual inspection methods are not only inefficient and labor-intensive, but also prone to inconsistent inspection standards due to factors such as personnel fatigue and subjective judgment, resulting in high rates of missed and false detections. These methods are no longer sufficient to meet the high-speed and high-precision requirements of modern automated production lines. However, existing defective product rejection structures equipped with AI vision scanning still suffer from limitations in effectively detecting defective products and are prone to misjudgments during inspection.

[0003] Therefore, it is essential to invent a defective product rejection structure with AI visual scanning. Utility Model Content

[0004] To address the aforementioned technical problems, this utility model provides a defective product rejection structure with AI visual scanning, solving the problem that existing defective product rejection structures with AI visual scanning still cannot effectively detect defective products and are prone to misjudgment during detection. A defective product rejection structure with AI visual scanning includes a conveyor belt, a conveying power structure, an AI vision device, a rejection structure, a transmission path, a return conveyor belt, and a return power structure. The conveying power structure is installed at one end of the conveyor belt, and the AI ​​vision device is fixedly installed on the outside of the feeding end of the conveyor belt. The rejection structure is fixedly installed on one side of the discharge end of the conveyor belt. The transmission path is fixedly installed on the other side of the discharge end of the conveyor belt, and the return conveyor belt is located below one side of the conveyor belt. The return power structure is installed at one end of the return conveyor belt.

[0005] The AI ​​vision device includes a support frame, a primary detection frame, a primary detection structure, an extension bracket, a follower push rod, a connecting frame, a follower detection frame, and a follower detection structure. The support frame is fixedly installed on the outer side of the feeding end of the conveyor belt, and the primary detection frame is fixedly installed on the inner side of the support frame. The primary detection structure is fixedly installed on the inner wall of the primary detection frame. The extension bracket is fixedly installed on the outer side of the support frame, and the follower push rod is fixedly installed inside the extension bracket. The connecting frame is fixedly installed at one end of the follower push rod. The follower detection frame is fixedly installed on the connecting frame, and the follower detection structure is fixedly installed on the inner wall of the follower detection frame.

[0006] The primary detection frame and primary detection structure inside the AI ​​vision device are fixed, while the follower detection frame and follower detection structure can reciprocate horizontally by following the push rod; the follower detection frame and follower detection structure can follow the defective product at a uniform speed; the primary detection frame and follower detection frame adopt two sets of "U"-shaped aluminum alloy frames; the primary detection structure and follower detection structure each adopt three sets, located at the top and sides of the frame respectively, and the primary detection structure and follower detection structure adopt high-definition cameras based on AI algorithms.

[0007] The internal structure of the return conveyor belt is exactly the same as that of the transport conveyor belt, and the transport direction of the return conveyor belt is opposite to that of the transport conveyor belt; the return conveyor belt can receive defective products falling from the transmission path and transport them back.

[0008] Compared with the prior art, the present invention has the following beneficial effects: The AI ​​vision device of this utility model performs preliminary image acquisition and intelligent analysis on the products on the conveyor belt through its fixed primary detection frame and primary detection structure, so as to achieve rapid screening. Once a suspected defective product is found, the follower push rod on its extension bracket drives the connecting frame, the follower detection frame and the follower detection structure to move synchronously with the target, so as to perform dynamic multi-angle accurate verification and final confirmation, and provide accurate decision signals for the execution of the rejection structure.

[0009] The present invention features a return conveyor belt that receives and transports defective products that have slipped off the transmission path, been identified by AI vision equipment, and rejected by the rejection structure. Driven by the return power structure, the defective products are returned to the loading end in the opposite direction to the transport conveyor belt, thereby completing the automatic recycling and centralized processing of defective products. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the structure of this utility model.

[0011] Figure 2 This is a structural schematic diagram of the AI ​​vision device of this utility model.

[0012] In the picture: 1. Conveyor belt; 2. Conveyor power structure; 3. AI vision device; 31. Support frame; 32. Primary detection frame; 33. Primary detection structure; 34. Extension bracket; 35. Follower push rod; 36. Connecting frame; 37. Follower detection frame; 38. Follower detection structure; 4. Rejection structure; 5. Conveyor path; 6. Return conveyor belt; 7. Return power structure. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0014] As attached Figure 1 To be continued Figure 2 As shown.

[0015] This utility model provides a defective product rejection structure with AI visual scanning, including a conveyor belt 1, a conveying power structure 2, an AI vision device 3, a rejection structure 4, a transmission path 5, a return conveyor belt 6, and a return power structure 7. The conveying power structure 2 is installed at one end of the conveyor belt 1, and the AI ​​vision device 3 is fixedly installed on the outside of the loading end of the conveyor belt 1. The rejection structure 4 is fixedly installed on one side of the discharge end of the conveyor belt 1. The transmission path 5 is fixedly installed on the other side of the discharge end of the conveyor belt 1, and the return conveyor belt 6 is located below one side of the conveyor belt 1. The return power structure 7 is installed at one end of the return conveyor belt 6.

[0016] The AI ​​vision device 3 includes a support frame 31, a primary detection frame 32, a primary detection structure 33, an extension bracket 34, a follower push rod 35, a connecting frame 36, a follower detection frame 37, and a follower detection structure 38. The support frame 31 is fixedly installed on the outer side of the feeding end of the conveyor belt 1, and the primary detection frame 32 is fixedly installed on the inner side of the support frame 31. The primary detection structure 33 is fixedly installed on the inner wall of the primary detection frame 32. The extension bracket 34 is fixedly installed on the outer side of the support frame 31, and the follower push rod 35 is fixedly installed inside the extension bracket 34. The connecting frame 36 is fixedly installed on one end of the follower push rod 35. The follower detection frame 37 is fixedly installed on the connecting frame 36, and the follower detection structure 38 is fixedly installed on the inner wall of the follower detection frame 37.

[0017] The primary detection frame 32 and primary detection structure 33 inside the AI ​​vision device 3 are fixed, while the follower detection frame 37 and follower detection structure 38 can reciprocate horizontally under the control of the follower push rod 35; the follower detection frame 37 and follower detection structure 38 can follow the defective product at a uniform speed; the primary detection frame 32 and follower detection frame 37 adopt two sets of "U"-shaped aluminum alloy frames; the primary detection structure 33 and follower detection structure 38 each adopt three sets, located at the top and sides of the frame respectively, and the primary detection structure 33 and follower detection structure 38 adopt high-definition cameras based on AI algorithms.

[0018] The internal structure of the return conveyor belt 6 is exactly the same as that of the transport conveyor belt 1, and the transport direction of the return conveyor belt 6 is opposite to that of the transport conveyor belt 1; the return conveyor belt 6 can receive defective products falling from the transmission path 5 and transport them back.

[0019] The overall working principle of this equipment begins with the conveyor power structure 2 driving the conveyor belt 1 to operate continuously, conveying the products to be inspected from the loading end to the unloading end. When the products pass through the fixedly installed AI vision equipment 3, they first enter its primary inspection frame 32 area, where the fixed primary inspection structure 33 performs rapid imaging and preliminary analysis based on AI algorithms to screen out suspected defective products.

[0020] Once a suspected target is identified, the tracking push rod 35 inside the AI ​​vision device 3 is immediately activated, pushing the tracking detection frame 37 and the internal tracking detection structure 38 to move synchronously with the product through the connecting frame 36. During this process, dynamic, multi-angle precise verification and final confirmation are performed.

[0021] Once a product is identified as defective, when it travels along conveyor belt 1 to the rejection structure 4 at the discharge end, the rejection structure 4 receives an instruction and pushes it away from the main line. The rejected defective product falls into the transmission path 5 under gravity and slides down onto the return conveyor belt 6 located below. Finally, driven by the return power structure 7, the return conveyor belt 6 transports the defective product back to the loading area in the opposite direction to conveyor belt 1, thus achieving automatic separation and recycling of good and defective products.

[0022] Any technical solution that achieves the above-mentioned technical effects by utilizing the technical solution described in this utility model, or by designing a similar technical solution inspired by the technical solution described in this utility model, falls within the protection scope of this utility model.

Claims

1. A defective product rejection structure with AI visual scanning, characterized in that: It includes a conveying conveyor belt (1), a conveying power structure (2), an AI vision device (3), a rejection structure (4), a conduction path (5), a return conveyor belt (6) and a return power structure (7), where: The conveying power structure (2) is installed at one end of the conveying conveyor belt (1), and the AI vision device (3) is fixedly installed outside the feeding end of the conveying conveyor belt (1), and the rejection structure (4) is fixedly installed on one side of the discharging end of the conveying conveyor belt (1); The conduction path (5) is fixedly installed on the other side of the discharging end of the conveying conveyor belt (1), and the return conveyor belt (6) is located below one side of the conveying conveyor belt (1), and the return power structure (7) is installed at one end of the return conveyor belt (6).

2. The defective product rejection structure with AI visual scanning as described in claim 1, characterized in that: The AI vision device (3) includes a support frame (31), a primary detection frame (32), a primary detection structure (33), an extension support (34), a following push rod (35), a connecting frame (36), a following detection frame (37) and a following detection structure (3), and the support frame (31) is fixedly installed outside the feeding end of the conveying conveyor belt (1), and the primary detection frame (32) is fixedly installed inside the support frame (31); The primary detection structure (33) is fixedly installed on the inner wall of the primary detection frame (32); The extension support (34) is fixedly installed outside the support frame (31), and the following push rod (35) is fixedly installed inside the extension support (34), and the connecting frame (36) is fixedly installed at one end of the following push rod (35); The following detection frame (37) is fixedly installed on the connecting frame (36), and the following detection structure (38) is fixedly installed on the inner wall of the following detection frame ().

3. The defective product rejection structure with AI visual scanning as described in claim 2, characterized in that: The primary detection frame (32) and the primary detection structure (33) inside the AI vision device (3) are fixed, and the following detection frame (37) and the following detection structure (38) can move reciprocally in the horizontal direction under the control of the following push rod (35); The following detection frame (37) and the following detection structure (38) can move at a constant speed following the defective products; The primary detection frame (32) and the following detection frame (37) adopt two sets of "冂"-shaped aluminum alloy frames; The primary detection structure (33) and the following detection structure (38) both adopt three groups, which are respectively located at the top and both sides of the frame, and the primary detection structure (33) and the following detection structure (38) adopt high-definition cameras based on AI algorithms.

4. The defective product rejection structure with AI visual scanning as described in claim 1, characterized in that: The internal structure of the return conveyor belt (6) is exactly the same as that of the conveying conveyor belt (1), and the conveying direction of the return conveyor belt (6) is opposite to that of the conveying conveyor belt (1); The return conveyor belt (6) can receive the defective products falling from the conduction path (5) and convey them back.

Images