An automated product inspection device based on visual recognition

CN224700593UActive Publication Date: 2026-09-01HUIZHOU RUICHENG ELECTRIC CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202522100203.4
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Filing Date
2025-09-28
Publication Date
2026-09-01
Estimated Expiration
2035-09-28

AI Technical Summary

Technical Problem

传统检测多依赖人工目视,存在效率低、劳动强度大、主观性强等问题,易因疲劳或经验差异导致漏检、误检,难以满足大规模量产的精度与效率需求

Benefits of technology

[0018] The automated product inspection device based on visual recognition of this utility model has at least one of the following beneficial effects during use:

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN224700593U_ABST
    Figure CN224700593U_ABST
Patent Text Reader

Abstract

The utility model discloses an automatic product detection device based on visual identification, including frame, the detection cabin of being located on frame, conveying belt, defective product recycling system and control system, the inside detection chamber of detection cabin is equipped with and connects the visual identification system of control system, the top of detection cabin has seted up and installed the window, be equipped with the detection station of corresponding visual identification system on the conveying belt, the conveying belt runs through detection cabin, photoelectric trigger sensor is provided with to the conveying belt at detection station. Whole process does not need manual intervention, from product delivery, photoelectric trigger detection, to image acquisition, analysis determination and defective product recycling are all automatic completion, effectively solve the problem that traditional manual detection is low, subjective, easy to fatigue, adapt to large -scale production line continuous detection demand, and the overall rhythm efficiency is greatly promoted.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This utility model relates to the field of detection device technology, specifically to an automated product detection device based on visual recognition. Background Technology

[0002] In the manufacturing process, product quality inspection is a crucial step in ensuring the quality of products leaving the factory. Traditional inspection methods rely heavily on manual visual inspection, which suffers from low efficiency, high labor intensity, and strong subjectivity. It is also prone to missed or false inspections due to fatigue or differences in experience, making it difficult to meet the accuracy and efficiency requirements of large-scale mass production.

[0003] While existing automated inspection equipment has partially replaced manual labor, it still has limitations: the light source design is simple, which easily produces reflections and shadows, making it difficult to clearly present subtle product defects; the position adjustment of the inspection camera is inconvenient, and it lacks flexibility when adapting to products of different sizes and shapes; the linkage between defective product sorting and inspection is insufficient, the degree of automation integration is low, and it affects the overall cycle time of the production line. Utility Model Content

[0004] In order to overcome the shortcomings of existing technical solutions, this utility model provides an automated product inspection device based on visual recognition, which can effectively solve the problems raised in the background technology.

[0005] The technical solution adopted by this utility model to solve its technical problem is:

[0006] An automated product inspection device based on vision recognition includes a frame, an inspection chamber mounted on the frame, a conveyor belt, a defective product recycling system, and a control system. The inspection chamber contains an inspection cavity and a vision recognition system connected to the control system. An installation window is provided on the top of the inspection chamber. The conveyor belt has inspection stations corresponding to the vision recognition system. The conveyor belt passes through the inspection chamber, and photoelectric trigger sensors are installed on the conveyor belt at the inspection stations.

[0007] The visual recognition system is fixedly installed on the top of the inspection chamber. The visual recognition system includes a three-dimensional adjustable bracket driven by a motor, an industrial camera installed on the three-dimensional adjustable bracket, and a multi-angle light source unit set on one side of the lens of the industrial camera. The optical axis of the industrial camera points to the product being tested at the inspection station.

[0008] The defective product recycling system includes a recycling chamber and a conveying mechanism mounted on a frame. The recycling chamber is fixedly located below the conveyor belt. The conveying mechanism includes a mounting base and a pneumatic clamping component mounted on the mounting base. The pneumatic clamping component is connected to the control system.

[0009] As a further description of the above technical solution, the control system is located outside the detection chamber. The control system includes a main controller and an image processing module. The main controller is electrically connected to and controls the conveyor belt, photoelectric trigger sensor, industrial camera and multi-angle light source unit.

[0010] As a further description of the above technical solution, the three-dimensional adjustable support includes an X-axis linear slide, a Y-axis linear slide, and a Z-axis lifting slide arranged perpendicularly to each other. The X-axis linear slide is connected to the Z-axis lifting slide, and the Y-axis linear slide is connected to the Z-axis lifting slide.

[0011] The industrial camera is fixed to the end of the X-axis lifting slide via a first quick-release locking module, and the multi-angle light source unit is fixed around the lens of the industrial camera via a second quick-release locking module. The multi-angle light source unit is composed of at least two LED light source modules with different illumination angles coaxially combined.

[0012] As a further description of the above technical solution, a coaxial auxiliary light source and a semi-transparent mirror are fixedly installed on the inner wall of the detection chamber directly above the detection station. The semi-transparent mirror is fixedly set at a 45° angle below the industrial camera lens.

[0013] As a further description of the above technical solution, a transparent observation window is embedded below the detection station, and a bottom supplement light that shines upward is correspondingly provided below the transparent observation window.

[0014] As a further description of the above technical solution, the inner wall of the detection chamber is covered with a light-absorbing material layer.

[0015] As a further description of the above technical solution, the signal output terminal of the industrial camera is connected to an image processing module, which is used to analyze the captured product image and output the detection result. The control system also includes a human-machine interface and an alarm.

[0016] As a further description of the above technical solution, the pneumatic clamping component includes a gripper, a telescopic arm, and a cylinder. The telescopic arm is connected to the gripper, and the cylinder drives the telescopic arm to reciprocate along both sides of the conveyor belt.

[0017] Compared with the prior art, the beneficial effects of this utility model are:

[0018] The automated product inspection device based on visual recognition of this utility model has at least one of the following beneficial effects during use:

[0019] The entire process requires no manual intervention. From product conveying and photoelectric trigger detection to image acquisition, analysis, and defective product recycling, everything is automated. This effectively solves the problems of low efficiency, strong subjectivity, and fatigue associated with traditional manual inspection, adapting to the continuous inspection needs of large-scale production lines and significantly improving overall cycle time. In terms of inspection accuracy, the combination of multi-angle light sources (including ring-shaped positive and oblique light sources, as well as coaxial and bottom supplementary lighting) combined with the light-absorbing material layer of the inspection chamber eliminates reflections, shadows, and ambient light interference, clearly highlighting subtle product defects and avoiding missed detections. The three-dimensional adjustable bracket combined with the quick-release locking module allows for precise adjustment of the industrial camera position, quickly adapting to products of different sizes and shapes without requiring the replacement of the entire equipment. Meanwhile, the control system's intuitive human-machine interface facilitates parameter setting and real-time monitoring, lowering the operational threshold. Defective product recycling and alarm linkage enable timely sorting of defective products, reducing quality risks. The quick-release design of core components also facilitates disassembly and calibration, reducing equipment downtime and ensuring stable operation. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall structure of an automated product inspection device based on vision recognition according to this utility model.

[0021] Figure 2 This is a top view of the automated product inspection device based on visual recognition according to this utility model.

[0022] Figure 3 This is a schematic diagram of the first part of the structure of an automated product inspection device based on vision recognition according to this utility model;

[0023] Figure 4 This is a schematic diagram of the second part of the structure of an automated product inspection device based on vision recognition according to this utility model;

[0024] Figure 5 This is a schematic diagram of the third part of an automated product inspection device based on visual recognition according to this utility model.

[0025] Numbering on the map:

[0026] 1. Frame; 101. Inspection chamber; 102. Conveyor belt; 103. Defective product recovery system; 104. Control system; 105. Photoelectric trigger sensor; 106. Recovery chamber; 107. Inspection station; 108. Transparent observation window; 2. Vision recognition system; 201. Industrial camera; 202. Multi-angle light source unit; 203. First quick-release locking module; 204. Second quick-release locking module; 3. Handling mechanism; 301. Mounting base; 302. Pneumatic clamping component; 303. Gripper; 304. Telescopic arm; 305. Cylinder; 4. Three-dimensional adjustable bracket; 401. X-axis linear slide; 402. Z-axis lifting slide; 403. Y-axis linear slide. Detailed Implementation

[0027] The technical solutions of the present utility model will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present utility model, and not all embodiments. Based on the embodiments of the present utility model, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present utility model.

[0028] like Figure 1-5 As shown, this utility model provides an automated product inspection device based on visual recognition, including a frame 1, an inspection chamber 101 mounted on the frame 1, a conveyor belt 102, a defective product recycling system 103, and a control system 104. The inspection chamber 101 is provided with an inspection cavity and a visual recognition system 2 connected to the control system 104. An installation window is provided on the top of the inspection chamber 101. The conveyor belt 102 is provided with an inspection station 107 corresponding to the visual recognition system 2. The conveyor belt 102 passes through the inspection chamber 101. A photoelectric trigger sensor 105 is provided on the conveyor belt 102 at the inspection station 107.

[0029] The product to be tested is conveyed to the testing station 107 inside the testing chamber 101 via conveyor belt 102. When the product arrives at the testing station 107, the photoelectric trigger sensor 105 preset on the conveyor belt 102 senses the product's position and immediately sends a trigger signal to the control system 104. After receiving the signal, the control system 104 controls the conveyor belt 102 to briefly pause or decelerate to lock the product's position and ensure that the product is in a stable state during testing; at the same time, it simultaneously starts the vision recognition system 2 and the light source unit to prepare for image acquisition.

[0030] The visual recognition system 2 is fixedly installed on the top of the inspection chamber 101. The visual recognition system 2 includes a three-dimensional adjustable bracket 4 driven by a motor, an industrial camera 201 installed on the three-dimensional adjustable bracket 4, and a multi-angle light source unit 202 set on one side of the lens of the industrial camera 201. The optical axis of the industrial camera 201 points to the product under test on the inspection station 107.

[0031] The photoelectric trigger sensor 105 is used to detect the position of the product under test at the detection station 107 and send a trigger signal to the control system 104. After receiving the trigger signal, the control system 104 controls the photoelectric trigger sensor 105 to lock the product position and controls the industrial camera 201 to acquire images.

[0032] The industrial camera 201 is mounted on a three-dimensional adjustable bracket 4 composed of linear slides in the X, Y, and Z directions. It can achieve precise spatial position adjustment through motor drive to ensure that the lens optical axis is vertically pointed to the product on the inspection station 107, adapting to the inspection needs of products of different sizes and shapes.

[0033] The multi-angle light source unit 202 (including the annular normal light source module and the annular oblique light source module with an included angle of 30°-60°) around the lens of the industrial camera 201 is started. Meanwhile, the coaxial auxiliary light source above the detection station 107 coincides with the optical path of the camera through the 45° half-transparent half-reflective mirror, and the bottom supplementary light lamp at the lower part irradiates upward through the transparent observation window 108. The multi-dimensional light source combination can eliminate interference such as reflection and shadow on the product surface, and highlight subtle defects (such as scratches, depressions, chromatic aberration, etc.). Under stable light source illumination, the industrial camera 201 collects high-definition images of the product, and the image data is transmitted to the image processing module of the control system 104 in real time.

[0034] The defective product recovery system 103 comprises a recovery chamber 106 arranged on the frame 1 and a conveying mechanism 3. The recovery chamber 106 is fixedly arranged below the conveyor belt 102, and the conveying mechanism 3 comprises a mounting seat 301 and a pneumatic clamping member 302 arranged on the mounting seat 301, and the pneumatic clamping member 302 is connected with the control system 104.

[0035] The image processing module performs algorithm analysis on the collected images (such as feature extraction, defect identification, size measurement, etc.), compares the analysis result with preset qualification standards, and outputs a "qualified" or "defective" detection result:

[0036] If the product is determined to be qualified, the control system 104 controls the conveyor belt 102 to continue running and conveys the product to the next process;

[0037] If the product is determined to be defective, the image processing module sends a signal to the main control machine, the main control machine triggers the alarm (sound and light alarm), and starts the defective product recovery system 103 at the same time.

[0038] To further explain, the control system 104 is arranged outside the detection cabin 101, and comprises a main control machine and an image processing module, and the main control machine is electrically connected to and controls the conveyor belt 102, the photoelectric trigger sensor 105, the industrial camera 201 and the multi-angle light source unit 202.

[0039] The combination of multi-angle light sources (normal light source, oblique light source, coaxial light source, bottom supplementary light) cooperating with the light-absorbing material layer on the inner wall of the detection cabin 101 can eliminate ambient light interference, highlight subtle defects on and inside the product surface, and solve the problem of missing detection caused by reflection and shadow under a single light source. The three-dimensional adjustable bracket 4 realizes precise adjustment of the spatial position of the industrial camera 201, and combined with the quick-install locking module, can quickly adapt to products of different sizes and shapes (such as electronic components, mechanical parts, plastic parts, etc.), without replacing the entire set of detection equipment.

[0040] Furthermore, the three-dimensional adjustable support 4 includes an X-axis linear slide 401, a Y-axis linear slide 403, and a Z-axis lifting slide 402 arranged perpendicularly to each other. The X-axis linear slide 401 is connected to the Z-axis lifting slide 402, and the Y-axis linear slide 403 is connected to the Z-axis lifting slide 402.

[0041] The industrial camera 201 is fixed to the end of the X-axis lifting slide via the first quick-release locking module 203. The multi-angle light source unit 202 is fixed around the lens of the industrial camera 201 via the second quick-release locking module 204. The multi-angle light source unit 202 is composed of at least two LED light source modules with different illumination angles coaxially combined.

[0042] It includes at least one annular positive light source module and one annular oblique light source module, wherein the center line of the LED emitting surface of the annular oblique light source module forms an angle of 30°-60° with the vertical direction. From product conveying, trigger detection, image acquisition to defective product recycling, the entire process requires no manual intervention, solving the problems of low efficiency, strong subjectivity, and easy fatigue in traditional manual inspection, and is suitable for the continuous inspection needs of large-scale production lines. The photoelectric trigger sensor 105 and the control system 104 have a rapid linkage response, which can realize "product arrival and inspection", with a short single product inspection cycle, improving the overall cycle efficiency of the production line.

[0043] Furthermore, a coaxial auxiliary light source and a semi-transparent mirror are fixedly installed on the inner wall of the detection chamber directly above the detection station 107. The semi-transparent mirror is fixed at a 45° angle below the lens of the industrial camera 201. The optical path of the coaxial auxiliary light source coincides with the optical path of the industrial camera 201 through the semi-transparent mirror.

[0044] Furthermore, a transparent observation window 108 is embedded below the detection station 107, and a bottom supplement light that shines upward is correspondingly provided below the transparent observation window 108.

[0045] Furthermore, the inner wall of the detection chamber 101 is covered with a light-absorbing material layer.

[0046] Furthermore, the signal output terminal of the industrial camera 201 is connected to an image processing module, which is used to analyze the captured product image and output the detection result. The control system 104 also includes a human-machine interface and an alarm.

[0047] The human-machine interface is used to set detection parameters, display real-time images and detection results; the alarm emits an alarm signal and / or controls the conveyor belt 102 to sort defective products when the image processing module determines that the product has a defect. Operators can preset detection parameters (such as defect threshold, light source brightness, etc.) through the human-machine interface of the control system 104, and view detection images, statistical data and alarm information in real time, realizing visual monitoring and flexible adjustment of the detection process.

[0048] The image processing module uses algorithms to objectively analyze images, avoiding subjective errors from manual inspection and ensuring high consistency of results. It also supports customizable parameters, allowing for flexible adjustment of inspection standards based on product characteristics. The alarm system is linked to the defective product recycling system 103 to ensure timely sorting of defective products, preventing them from flowing into subsequent processes and reducing quality risks.

[0049] The human-machine interface intuitively displays detection data and images, facilitating real-time monitoring and parameter adjustment by operators, thus lowering the operational threshold. Core components such as the industrial camera 201 and the light source unit are installed via quick-release locking modules, facilitating disassembly, calibration, and maintenance, and reducing equipment downtime.

[0050] Furthermore, the pneumatic gripper 302 includes grippers 303, a telescopic arm 304, and a cylinder 305. The telescopic arm 304 is connected to the grippers 303, and the cylinder 305 drives the telescopic arm 304 to reciprocate along both sides of the conveyor belt 102. The pneumatic gripper 302 on the mounting base 301 (driven by the cylinder 305 to move the telescopic arm 304 and thus the grippers 303) extends from both sides of the conveyor belt 102, precisely gripping defective products and then transferring them to the recycling chamber 106 below the conveyor belt 102 for sorting and recycling. After recycling, the gripper resets, and the conveyor belt 102 resumes normal operation.

[0051] It will be apparent to those skilled in the art that this invention is not limited to the details of the exemplary embodiments described above, and that it can be implemented in other specific forms without departing from the spirit or essential characteristics of this invention. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within this invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An automated product inspection device based on visual recognition, characterized in that: The system includes a frame, a testing chamber mounted on the frame, a conveyor belt, a defective product recycling system, and a control system. The testing chamber contains a testing room and a visual recognition system connected to the control system. The top of the testing chamber has an installation window. The conveyor belt has testing stations corresponding to the visual recognition system. The conveyor belt runs through the testing chamber. The conveyor belt is equipped with photoelectric trigger sensors at the testing stations. The visual recognition system is fixedly installed on the top of the inspection chamber. The visual recognition system includes a three-dimensional adjustable bracket driven by a motor, an industrial camera installed on the three-dimensional adjustable bracket, and a multi-angle light source unit set on one side of the lens of the industrial camera. The optical axis of the industrial camera points to the product being tested at the inspection station. The defective product recycling system includes a recycling chamber and a conveying mechanism mounted on a frame. The recycling chamber is fixedly located below the conveyor belt. The conveying mechanism includes a mounting base and a pneumatic clamping component mounted on the mounting base. The pneumatic clamping component is connected to the control system.

2. The automated product inspection device based on visual recognition according to claim 1, characterized in that: The control system is located outside the detection chamber. The control system includes a main controller and an image processing module. The main controller is electrically connected to and controls the conveyor belt, photoelectric trigger sensor, industrial camera and multi-angle light source unit.

3. The automated product inspection device based on visual recognition according to claim 1, characterized in that: The three-dimensional adjustable support includes an X-axis linear slide, a Y-axis linear slide, and a Z-axis lifting slide arranged perpendicularly to each other. The X-axis linear slide is connected to the Z-axis lifting slide, and the Y-axis linear slide is connected to the Z-axis lifting slide. The industrial camera is fixed to the end of the X-axis lifting slide via a first quick-release locking module, and the multi-angle light source unit is fixed around the lens of the industrial camera via a second quick-release locking module. The multi-angle light source unit is composed of at least two LED light source modules with different illumination angles coaxially combined.

4. The automated product inspection device based on visual recognition according to claim 1, characterized in that: A coaxial auxiliary light source and a semi-transparent mirror are fixedly installed on the inner wall of the detection chamber directly above the detection station. The semi-transparent mirror is fixed at a 45° angle below the industrial camera lens.

5. The automated product inspection device based on visual recognition according to claim 1, characterized in that: A transparent observation window is embedded below the testing station, and a bottom supplement light that shines upwards is correspondingly installed below the transparent observation window.

6. The automated product inspection device based on visual recognition according to claim 1, characterized in that: The inner wall of the testing chamber is covered with a layer of light-absorbing material.

7. The automated product inspection device based on visual recognition according to claim 2, characterized in that: The industrial camera signal output terminal is connected to the image processing module, which is used to analyze the captured product image and output the detection result. The control system also includes a human-machine interface and an alarm.

8. The automated product inspection device based on visual recognition according to claim 1, characterized in that: The pneumatic clamping component includes a gripper, a telescopic arm, and a cylinder. The telescopic arm is connected to the gripper, and the cylinder drives the telescopic arm to reciprocate along both sides of the conveyor belt.