Visual identification and classification mechanism for feature defects of parts
By designing a visual recognition and classification mechanism for part feature defects, and utilizing components such as imaging cameras, lighting lamps, and alarms, rapid and accurate part inspection is achieved, solving the problems of slow speed and unstable accuracy of traditional manual inspection, and meeting the needs of large-scale production.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-07
- Publication Date
- 2026-04-03
AI Technical Summary
Traditional manual visual inspection of parts results in slow inspection speed, making it difficult to meet the needs of large-scale production. The inspection accuracy is also unstable, and it is easy to miss minor defects.
Design a visual recognition and classification mechanism for part feature defects. It adopts an image acquisition camera, a lighting lamp, an alarm and a controller, combined with an image collection and comparison component, a reflection component and a clamping component to achieve fast and accurate acquisition and analysis of part images, identify defects through algorithms and issue timely alarms.
It enables high-speed online inspection of parts, improves inspection efficiency and accuracy, ensures the accuracy and stability of inspection results, reduces the influence of human factors, and improves the repeatability and consistency of inspection.
Smart Images

Figure CN224081470U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of visual recognition technology, specifically a visual recognition and classification mechanism for part feature defects. Background Technology
[0002] Parts are the basic units that make up machines or structures. Various abnormal situations that occur during the manufacturing, processing or use of parts that do not meet design requirements or affect their performance and reliability, and the characteristic defects of parts may affect the assembly accuracy, service life, reliability of parts and the performance of the entire equipment. Therefore, in the production and quality inspection of parts, effective measures need to be taken to detect and control these defects to ensure that the quality and performance of parts meet the requirements.
[0003] Traditional manual visual inspection of parts requires inspectors to concentrate on observing the parts for extended periods, which can easily lead to visual fatigue and slow inspection speed. In large-scale production environments, it is difficult to meet the needs of efficient production. Long hours of repetitive work can also reduce the inspector's attention, thus affecting the accuracy of the inspection and making it easy to miss some minor defects, resulting in unstable inspection precision. Utility Model Content
[0004] The purpose of this invention is to provide a visual identification and classification mechanism for part feature defects, in order to solve the problem mentioned in the background art that the traditional manual visual inspection of parts requires the inspector to concentrate on observing the parts for a long time, which easily leads to visual fatigue and slow inspection speed. In a large-scale production environment, it is difficult to meet the needs of efficient production. Long-term repetitive work will reduce the inspector's attention, thereby affecting the accuracy of inspection and making it easy to miss some minor defects, resulting in unstable inspection accuracy.
[0005] To achieve the above objectives, this utility model provides the following technical solution: a visual recognition and classification mechanism for part feature defects, comprising a mounting base, fixed vertical plates fixedly connected to the top of both sides of the mounting base, fixed rotating shafts rotatably connected to the top of each of the two fixed vertical plates, a rotating frame fixedly connected between the two fixed rotating shafts, a drive motor fixedly mounted on the side of one of the fixed vertical plates, an image collection and comparison component fixedly mounted in the middle of the rotating frame, a reflection component symmetrically fixedly connected to the inner sidewall of the rotating frame, a support base fixedly connected to the top center of the mounting base, a support frame fixedly connected to the top center of the support base, a pass-through notch on the top of the support frame, and clamping components symmetrically fixedly connected to the top of the support frame. This device can quickly acquire and analyze part images, enabling high-speed online detection of parts, improving detection efficiency, meeting the needs of large-scale production, detecting defects at smaller scales, and achieving detection accuracy far exceeding that of manual visual inspection. It can effectively ensure part quality, is unaffected by human factors such as fatigue or emotions, and provides accurate, stable, and highly repeatable and consistent detection results.
[0006] Preferably, the image collection and comparison component includes an image acquisition camera, which is fixedly embedded in the top of the inner side of the rotating frame. A lighting lamp is symmetrically fixedly installed on the side of the image acquisition camera, and an alarm is fixedly connected to the other corner of the image acquisition camera. With the image acquisition camera and lighting lamps, when the detection result shows that the part has a defect, the alarm will sound in time, which makes it convenient for the staff to remove the part and classify it.
[0007] Preferably, the reflective assembly includes two fixed frames, which are symmetrically fixedly connected to the inner wall of the rotating frame. A refractive lens is fixedly connected inside the fixed frame. The refractive lens can refract and converge the light from the part, so that the part forms a clear image on the imaging camera.
[0008] Preferably, the clamping assembly includes two fixed upright plates, which are symmetrically fixedly connected to the top of the support frame. Each of the two fixed upright plates has a sliding rod passing through it. An anti-detachment block is fixedly connected to the ends of the two sliding rods that are far apart from each other. A limit ring is fixedly connected to the middle of each of the two sliding rods. A top spring is movably connected between the limit ring and the anti-detachment block. Support columns are symmetrically fixedly connected to the sides of the two fixed upright plates that are far apart from each other. A connecting plate is fixedly connected to the end of the support column that is far from the fixed upright plate. The sliding rod passes through the middle of the connecting plate. An anti-detachment clamping block is fixedly connected to the ends of the two sliding rods that are close to each other. The clamping assembly clamps the part, keeping it in a relatively stable state, facilitating multi-angle observation of the part.
[0009] Preferably, the fixed upright plate has a circular through hole in the middle, and a limit rod is symmetrically fixedly connected inside the circular through hole. A limit groove is symmetrically opened on the outer side of the sliding rod. The limit groove and the limit rod match to prevent the sliding rod from rotating during the sliding process, which would affect the clamping stability.
[0010] Preferably, the mounting base has symmetrical circular fixing holes in the middle, and a controller is fixedly connected to the side of the fixing plate. The controller is used to collect and process images and react according to the image information to make the alarm work.
[0011] Compared with the prior art, the beneficial effects of this utility model are:
[0012] The system consists of an image camera, a lighting lamp, an alarm, and a controller. The controller receives image data from the image camera and analyzes the images using preset algorithms and models. When the controller determines that an abnormality has been detected based on the image analysis results, it sends an alarm signal to the alarm. Upon receiving the signal, the alarm emits an alarm through its built-in sound, light, or vibration device, prompting staff to remove the inspected parts, classify and place them, and prevent defective parts from being mixed in, which would affect the yield rate of the parts processing.
[0013] The design incorporates a fixed frame, refractive lens, drive motor, support base, support frame, through notch, fixed upright plate, sliding rod, limit rod, limit groove, support column, connecting plate, anti-detachment block, limit retaining ring, top spring, and anti-detachment clamping block, which can quickly clamp parts and improve inspection efficiency. At the same time, it can drive the imaging camera to rotate, allowing for observation of parts at a wider angle, thus facilitating faster inspection. Attached Figure Description
[0014] Figure 1 This is a three-dimensional structural diagram of the present invention;
[0015] Figure 2 This is a top view of the present invention;
[0016] Figure 3 This utility model Figure 1 A magnified view of the structure at point A in the middle;
[0017] Figure 4 This is a schematic diagram showing the positional distribution of the fixing frame and refractive lens of this utility model.
[0018] In the diagram: 1. Mounting base; 2. Fixed vertical plate; 3. Fixed rotating shaft; 4. Rotating frame; 5. Image capture camera; 6. Lighting lamp; 7. Alarm; 8. Fixing frame; 9. Reflecting lens; 10. Drive motor; 11. Controller; 12. Support base; 13. Support frame; 14. Through notch; 15. Fixed vertical plate; 16. Sliding rod; 17. Limiting rod; 18. Limiting groove; 19. Support column; 20. Connecting plate; 21. Anti-detachment block; 22. Limiting retaining ring; 23. Top spring; 24. Anti-detachment clamping block. Detailed Implementation
[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention.
[0020] Please see Figure 1-4 This utility model provides a visual recognition and classification mechanism for part feature defects, including a mounting base 1, with fixed vertical plates 2 fixedly connected to the top of both sides of the mounting base 1, fixed rotating shafts 3 rotatably connected to the top of each of the two fixed vertical plates 2, and a rotating frame 4 fixedly connected between the two fixed rotating shafts 3. A drive motor 10 is fixedly installed on the side of one of the fixed vertical plates 2, and an image collection and comparison component is fixedly installed in the middle of the rotating frame 4, mainly used for image acquisition and image transmission. A reflection component is symmetrically fixedly connected to the inner side wall of the rotating frame 4, so that the image of the side of the part can also be presented in front of the lens of the image acquisition camera 5 after reflection, which facilitates the acquisition of images of the part at a larger angle. A support base 12 is fixedly connected to the top middle of the mounting base 1, and a support frame 13 is fixedly connected to the top middle of the support base 12. A passage notch 14 is opened on the top of the support frame 13, and clamping components are symmetrically fixedly connected to the top of the support frame 13.
[0021] Furthermore, the image acquisition and comparison component includes an image capture camera 5, which has the ability to quickly and accurately capture images of the object being inspected, convert the optical information of the object into electrical signals, and then generate digital images. It is capable of detecting minute size changes and defects. The image capture camera 5 is fixedly embedded in the top of the inner side of the rotating frame 4. Illumination lamps 6 are symmetrically fixedly installed on the side of the image capture camera 5. The illumination lamps 6 are used to provide a stable lighting environment for the image capture camera 5, which facilitates the extraction of clearer images. An alarm 7 is fixedly connected to the corner of the other side of the image capture camera 5. The alarm 7 is used to sound an alarm when abnormalities occur in the part features, prompting the staff to remove the part and classify and place it.
[0022] Furthermore, the reflective assembly includes two fixed frames 8, which are symmetrically fixed to the inner wall of the rotating frame 4. A refractive lens 9 is fixedly connected inside the fixed frame 8 to reflect the image of the side of the part, so that the imaging camera 5 can capture the image of the side of the part.
[0023] Furthermore, the clamping assembly includes two fixed upright plates 15, which are symmetrically fixedly connected to the top of the support frame 13. Each of the two fixed upright plates 15 has a sliding rod 16 passing through it. Anti-detachment blocks 21 are fixedly connected to the ends of the two sliding rods 16 that are furthest from each other. Limiting rings 22 are fixedly connected to the middle of each of the two sliding rods 16. A top spring 23 is movably connected between the limiting rings 22 and the anti-detachment blocks 21. Support columns 19 are symmetrically fixedly connected to the sides of the two fixed upright plates 15 that are furthest from each other. A connecting... Plate 20, sliding rod 16 passes through the middle of connecting plate 20, and anti-detachment clamping block 24 is fixedly connected to the end of the two sliding rods 16 that are close to each other. When installing the part, the part is placed between the two anti-detachment clamping blocks 24. During this process, the anti-detachment clamping block 24 pushes the sliding rod 16 to slide in the fixed upright plate 15, the limiting rod 17 and the limiting groove 18 move relative to each other, and at the same time the limiting stop ring 22 pushes the top spring 23 to compress, so that the sliding rod 16 and the connecting plate 20 slide relative to each other. Then the reaction force of the top spring 23 is used to clamp the part. This device is used for parts with small mass.
[0024] Furthermore, a circular through hole is provided in the middle of the fixed plate 15, and a limit rod 17 is symmetrically fixedly connected inside the circular through hole. A limit groove 18 is symmetrically provided on the outer side of the sliding rod 16, and the limit groove 18 matches the limit rod 17.
[0025] Furthermore, the mounting base 1 has symmetrical circular fixing holes in the middle, and the side of the fixing plate 2 is fixedly connected to the controller 11. The controller 11 is responsible for coordinating the work of each component. When the system starts, the controller 11 initializes the camera 5, the lighting lamp 6 and other devices, and sets their working parameters, such as the exposure time and frame rate of the camera 5, and the brightness and color of the lighting lamp 6. The controller 11 receives image data from the camera 5 and uses preset algorithms and models to analyze and process the image. These algorithms can identify the shape, size, position, color and other features of the object, and detect whether there are defects or abnormalities.
[0026] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A mechanism for visual recognition and classification of defects of features of a part, comprising a mounting base (1), characterized in that: Both sides of the mounting base (1) top end are fixedly connected with fixed vertical plate (2), two fixed vertical plate (2) top end are rotatably connected with fixed rotating shaft (3), two fixed rotating shaft (3) are fixedly connected with rotating frame (4), one fixed vertical plate (2) side surface is fixedly installed with drive motor (10), the rotating frame (4) middle part is fixedly installed with image collection contrast assembly, the rotating frame (4) inner side wall is fixedly connected with reflection assembly, the mounting base (1) top end middle part is fixedly connected with support base (12), the support base (12) top end middle part is fixedly connected with support frame (13), the support frame (13) top is provided with through gap (14), the support frame (13) top end is fixedly connected with clamping assembly.
2. The part feature defect visual recognition and classification mechanism according to claim 1, characterized in that: The image collection contrast assembly includes an image camera (5), the image camera (5) is fixedly embedded in the rotating frame (4) inner side surface top, the image camera (5) side surface is fixedly installed with illuminating lamp (6), the other side edge position of the image camera (5) is fixedly connected with alarm (7).
3. The part feature defect visual recognition and classification mechanism according to claim 1, characterized in that: The reflection assembly includes two fixed frames (8), two fixed frames (8) are fixedly connected to the rotating frame (4) inner side wall, the fixed frame (8) is fixedly connected with refractive lens (9) inside.
4. The part feature defect visual recognition and classification mechanism of claim 1, wherein: The clamping assembly includes two fixed vertical plates (15), two fixed vertical plates (15) are fixedly connected to the support frame (13) top end, two sliding rods (16) are provided in two fixed vertical plates (15), two sliding rods (16) are fixedly connected with anti-drop block (21) at one end away from each other, two sliding rods (16) are fixedly connected with limiting stop ring (22) in the middle, the limiting stop ring (22) and anti-drop block (21) are movably connected with top spring (23), two fixed vertical plates (15) are fixedly connected with support column (19) on one side away from each other, the support column (19) is fixedly connected with connecting plate (20) at one end away from the fixed vertical plate (15), the sliding rod (16) penetrates the middle part of the connecting plate (20), two sliding rods (16) are fixedly connected with anti-drop clamping block (24) at one end close to each other.
5. The part feature defect visual recognition and classification mechanism according to claim 4, characterized in that: The fixed vertical plate (15) is provided with a circular through hole in the middle, the circular through hole is fixedly connected with limiting rod (17) inside, the sliding rod (16) is provided with limiting slot (18) symmetrically outside, the limiting slot (18) and limiting rod (17) are matched.
6. The part feature defect visual recognition and classification mechanism of claim 1, wherein: The mounting base (1) is provided with circular fixing hole symmetrically in the middle, the fixed vertical plate (2) is fixedly connected with controller (11) on the side surface.