Object sorting device based on visual identification

The visual recognition device, which combines a material tray and a multi-axis electric cylinder, solves the problems of low sorting efficiency and high motor wear in the existing technology, and achieves efficient sorting of winter jujubes and rejection of defective products.

CN223733331UActive Publication Date: 2025-12-30SU ZHOU ZHENG HAO ZHI NENG KE JI YOU XIAN GONG SI
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202423101012.1
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Filing Date
2024-12-16
Publication Date
2025-12-30
Estimated Expiration
2034-12-16

AI Technical Summary

Technical Problem

Existing vision-based automatic sorting devices have low sorting efficiency and high stepper motor operating losses. They also have difficulty in achieving quantitative and intermittent material drop detection, resulting in poor sorting performance for winter jujubes.

Method used

The system combines a feeding tray, a vibration assembly, a high-definition camera, a laser emitter, and a multi-axis electric cylinder. It disperses the jujube grains through vibration and establishes a coordinate system. The high-definition camera analyzes the quality of the jujubes, and the main telescopic electric cylinder and clamps perform precise sorting, thus achieving quantitative sorting.

Benefits of technology

It improves sorting efficiency, reduces the operating losses of stepper motors, ensures that each date has a corresponding coordinate point, and achieves efficient quantitative sorting and rejection of defective products.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN223733331U_ABST
    Figure CN223733331U_ABST
Patent Text Reader

Abstract

The article sorting device comprises a sorting table, a material containing disc and a vibration assembly used for driving the material containing disc to vibrate are horizontally arranged in the middle of the upper portion of the sorting table, a top plate is fixedly arranged over the sorting table, and a main telescopic electric cylinder is movably hinged to the middle of the bottom of the top plate in a universal mode. A piston rod and a clamp located at the tail end of the piston rod are arranged at the output end of the bottom of the main telescopic electric cylinder, light spots are vertically emitted through the laser emitter, and the controller calculates and establishes a material containing disc with a coordinate system according to the two point positions and a shot picture, so that each jujube grain has the corresponding coordinate point position; meanwhile, the controller can analyze and judge the positions and the number of the winter jujube grains with poor conditions according to detail picture information fed back by a plurality of high-definition cameras, and sort and remove the bad grains under the cooperation of a main telescopic electric cylinder, an X-axis adjusting electric cylinder, a Y-axis adjusting electric cylinder, a clamp and the like, and all the winter jujube grains do not need to be recognized and sorted one by one.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This utility model relates to the field of sorting equipment technology, specifically to an item sorting device based on visual recognition. Background Technology

[0002] Machine vision is a rapidly developing branch of artificial intelligence that uses machines to replace human eyes for measurement and judgment. A machine vision system uses an image acquisition device to convert the captured target into an image signal, which is then transmitted to a dedicated image processing system. This system obtains the target's shape information and, based on pixel distribution, brightness, color, and other information, converts it into a digital signal. The image system performs various calculations on these signals to extract the target's features, such as area, quantity, position, and length. Based on preset tolerances and other conditions, it outputs results, including size, angle, number, pass / fail status, presence / absence, etc., achieving automatic recognition. Finally, based on the judgment results, it controls the equipment's movement.

[0003] Currently, the sorting of winter jujubes by appearance and quality is done using color sorters. Patent document CN210816317U discloses an automatic sorting device based on visual recognition. This device uses existing visual recognition sensors to select winter jujubes by color, and the sorting wheels achieve both sorting and cost savings. However, this device has significant shortcomings in practical use. For example, during sorting, all winter jujubes need to be individually identified and sorted, while the vast majority of winter jujubes are of acceptable appearance and require no inspection, resulting in significantly low sorting efficiency and a substantial increase in the operating wear of the stepper motor. Furthermore, this automatic sorting device struggles to achieve quantitative, intermittent material detection during actual use, leading to the problem of multiple winter jujubes being discharged from the storage bin at once, thus affecting the sorting effect. Therefore, a visual recognition-based item sorting device is designed. Utility Model Content

[0004] The purpose of this section is to outline some aspects of the embodiments of this utility model and to briefly introduce some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be used to limit the scope of this utility model.

[0005] Therefore, the purpose of this utility model is to provide a visual recognition-based item sorting device to solve the problems mentioned in the background art, such as the low sorting efficiency of the existing visual recognition-based automatic sorting device, which also significantly increases the operating loss of the stepper motor; on the other hand, the automatic sorting device is difficult to achieve quantitative intermittent material dropping detection in actual use, and multiple winter dates are easily discharged from the storage box at one time, thus affecting the sorting effect.

[0006] To achieve the above objectives, this utility model provides the following technical solution: a visual recognition-based item sorting device, comprising a sorting table, a material tray horizontally arranged at the center of the upper part of the sorting table and a vibration component for driving the material tray to vibrate, a top plate fixedly arranged directly above the sorting table, a main telescopic electric cylinder omnidirectionally hinged at the center of the bottom of the top plate, a piston rod and a clamp located at the end of the piston rod at the bottom output end of the main telescopic electric cylinder, and an angle adjustment component for driving the main telescopic electric cylinder to swing and rotate circumferentially, a mounting plate fixed to the lower outer periphery of the main telescopic electric cylinder, a plurality of high-definition cameras distributed around the bottom outer periphery of the mounting plate, and a central control unit, i.e., a controller, for the sorting device.

[0007] As a preferred embodiment of the visual recognition-based item sorting device of this utility model, the top surface of the material tray has several closely adjacent positioning grooves, and the outer periphery of the top of the material tray has a baffle plate.

[0008] As a preferred embodiment of the visual recognition-based item sorting device of this utility model, the vibration component includes a plurality of spring legs arranged radially around the bottom of the material tray and a vibration motor installed at the center of the bottom of the material tray, and the lower ends of the spring legs are fixedly installed on the top of the sorting table.

[0009] As a preferred embodiment of the visual recognition-based item sorting device of this utility model, the angle adjustment component includes an X-axis adjustment cylinder installed at the bottom of the top plate and with its output end connected to one side of the main telescopic cylinder, a Y-axis adjustment cylinder installed at the bottom of the top plate and with its output end connected to the other side of the main telescopic cylinder, and universal joints disposed at the beginning and end of the X-axis adjustment cylinder, the beginning and end of the Y-axis adjustment cylinder, and the root of the top of the main telescopic cylinder.

[0010] As a preferred embodiment of the visual recognition-based item sorting device of this utility model, at least two laser emitters connected to the controller signal are also provided at the bottom of the top plate, and the laser beams emitted by the laser emitters are perpendicular to the material tray.

[0011] As a preferred embodiment of the visual recognition-based item sorting device described in this utility model, the top of the sorting table also has a recessed annular waste collection groove, which is distributed around the outer diameter of the material tray.

[0012] As a preferred embodiment of the visual recognition-based item sorting device described in this utility model, the sorting platform also has a clearance groove corresponding to the vibration motor at its center.

[0013] As a preferred embodiment of the visual recognition-based item sorting device described in this utility model, the top of the sorting table is further provided with a support frame for fixing the top plate, and the controller is installed on one side of the support frame.

[0014] Compared with existing technologies, the advantages of this invention are as follows: This visual recognition-based sorting device directly pours a fixed quantity of dates onto a feeding tray during sorting. Then, under the action of the vibration component, the granules on the tray quickly roll and disperse, falling into their corresponding positioning grooves. This creates a certain gap between adjacent granules, preventing the drawback of difficulty in identification and sorting due to accumulation. Subsequently, a laser emitter vertically emits light spots, and the controller calculates and establishes a coordinate system for the feeding tray based on two points and the captured image, ensuring that each date has a corresponding coordinate point. Simultaneously, the controller analyzes and judges the location and quantity of inferior dates based on detailed image information from multiple high-definition cameras. With the cooperation of the main telescopic electric cylinder, X-axis adjusting electric cylinder, Y-axis adjusting electric cylinder, and clamps, defective granules are sorted and removed. This eliminates the need for individual identification and sorting of all winter dates, significantly improving sorting efficiency. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall structure of this utility model;

[0016] Figure 2 This is a schematic diagram of the vibration component and the material tray of this utility model;

[0017] Figure 3 This is a schematic diagram of the angle adjustment component of this utility model.

[0018] In the diagram: 100, sorting table; 110, annular waste collection trough; 120, clearance trough; 200, material tray; 210, positioning groove; 220, baffle plate; 300, vibration assembly; 310, spring leg; 320, vibration motor; 400, top plate; 410, support frame; 500, main telescopic electric cylinder; 501, piston rod; 510, angle adjustment assembly; 511, X-axis adjustment electric cylinder; 512, Y-axis adjustment electric cylinder; 513, universal joint; 600, clamp; 700, mounting plate; 710, high-definition camera; 800, laser emitter; 900, controller. Detailed Implementation

[0019] To make the above-mentioned objectives, features and advantages of this utility model more apparent and understandable, the specific embodiments of this utility model will be described in detail below with reference to the accompanying drawings.

[0020] Secondly, this utility model is described in detail with reference to the schematic diagrams. When describing the embodiments of this utility model, for ease of explanation, the cross-sectional views showing the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of this utility model. In addition, in actual manufacturing, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0021] To make the objectives, technical solutions, and advantages of this utility model clearer, the embodiments of this utility model will be described in further detail below with reference to the accompanying drawings.

[0022] Figures 1-3 The diagram shown is a complete structural schematic of a visual recognition-based item sorting device according to this utility model. Please refer to [link / reference]. Figures 1-3 This embodiment of a visual recognition-based item sorting device includes a sorting table 100, a material tray 200 and a vibration component 300 for driving the material tray 200 to vibrate, a top plate 400 fixedly installed directly above the sorting table 100, a main telescopic electric cylinder 500 omnidirectionally hinged at the bottom center of the top plate 400, a piston rod 501 and a clamp 600 located at the end of the piston rod 501 at the bottom output end of the main telescopic electric cylinder 500, an angle adjustment component 510 for driving the main telescopic electric cylinder 500 to swing and rotate circumferentially, a mounting plate 700 fixedly installed on the lower outer periphery of the main telescopic electric cylinder 500, a plurality of high-definition cameras 710 distributed around the bottom outer periphery of the mounting plate 700, and a general control unit, i.e., a controller 900, for the sorting device.

[0023] The top surface of the feeding tray 200 has several closely adjacent positioning grooves 210, and the outer periphery of the top of the feeding tray 200 has a baffle plate 220. It can be understood that the positioning grooves 210 are used to position and stabilize the corresponding jujube grains, preventing them from rolling randomly during sorting and removal. Furthermore, the spacing between the jujube grains allows the high-definition camera 710 to more clearly capture the appearance of the outer surface. It is important to note that the number of jujubes after dispersion should not exceed the number of positioning grooves 210 each time material is added, to avoid accumulation and interference with identification and sorting. In addition, the baffle plate 220 prevents the granules from rolling out of the feeding tray 200 during vibration dispersion. The vibration assembly 300 includes multiple spring legs 310 arranged radially around the bottom of the feeding tray 200, and a vibration motor 320 installed at the center of the bottom of the feeding tray 200. The lower ends of the spring legs 310 are fixedly installed on the top of the sorting table 100. The angle adjustment assembly 510 includes an X-axis adjusting cylinder 511 mounted on the bottom of the top plate 400 with its output end connected to one side of the main telescopic cylinder 500; a Y-axis adjusting cylinder 512 mounted on the bottom of the top plate 400 with its output end connected to the other side of the main telescopic cylinder 500; and a universal joint 513 located at the beginning and end of the X-axis adjusting cylinder 511, the beginning and end of the Y-axis adjusting cylinder 512, and the root of the top of the main telescopic cylinder 500. It can be understood that through the telescopic cooperation of the X-axis adjusting cylinder 511, the Y-axis adjusting cylinder 512, and the universal joint 513, the output end of the main telescopic cylinder 500 can point to any point on the material tray 200 during rotational adjustment, which is beneficial for the quick sorting and rejection of jujubes. At least two laser emitters 800 connected to the controller 900 are also provided at the bottom of the top plate 400, and the laser beam emitted by the laser emitters 800 is perpendicular to the material tray 200. It is understandable that the main telescopic electric cylinder 500, which drives the clamp 600 to grab defective dates, needs a precise positioning coordinate point. Therefore, the material tray 200 here is equivalent to a coordinate platform. When at least two laser emitters 800 emit light onto the material tray 200, they represent two points in the coordinate system. That is, two points determine the coordinate system. In this way, the controller 900 will calculate and establish a material tray 200 with a coordinate system based on the two points and the image captured by the camera, so that each date has a corresponding coordinate point. Only in this way can the main telescopic electric cylinder 500 and the X-axis adjusting electric cylinder 511 be precisely adjusted and grabbed.Specifically, in this embodiment, during sorting, a fixed quantity of dates is poured directly onto the feeding tray 200. Then, under the action of the vibration component 300, the granules on the tray surface are quickly rolled and dispersed, falling into the corresponding positioning grooves 210. In this way, there is a certain gap between adjacent granules, preventing the disadvantage of being difficult to identify and sort due to accumulation. Subsequently, the laser emitter 800 emits light spots vertically, and the controller 900 calculates and establishes a feeding tray 200 with a coordinate system based on the two points and the captured image, so that each date has a corresponding coordinate point. At the same time, the controller 900 analyzes and judges the position and quantity of date granules with poor appearance based on the detailed image information fed back by multiple high-definition cameras 700, and sorts and removes defective granules with the cooperation of the main telescopic electric cylinder 500, the X-axis adjusting electric cylinder 511, the Y-axis adjusting electric cylinder 512 and the clamp 600, etc., without having to identify and sort all the winter date granules one by one.

[0024] It should be noted that in other embodiments, the feeding tray 200 can be configured as a detachable connection, that is, after the date grains in the feeding tray 200 are identified and sorted, the feeding tray 200 can be removed and the granules poured out, and then the next identification and sorting can be performed. There is no limitation on this.

[0025] Preferably, the top of the sorting table 100 also has a recessed annular waste collection trough 110, which is distributed around the outer diameter of the material tray 200. The annular waste collection trough 110 allows the clamp 600 to easily discard defective dates into the adjacent annular waste collection trough 110 during the sorting process, improving sorting efficiency while preventing contamination caused by the indiscriminate discarding of defective dates.

[0026] Preferably, the sorting table 100 also has a clearance groove 120 corresponding to the vibrating motor 320 at its center. The clearance groove 120 can prevent the vibrating motor 320 from colliding with the top of the sorting table 100 during operation.

[0027] Preferably, the top of the sorting table 100 also has a support frame 410 for fixing the top plate 400 around its perimeter, and the controller 900 is installed on one side of the support frame 410. The support frame 410 can be used to position and install the top plate 400 and the sorting mechanism, so that the sorting device forms an integral equipment platform.

[0028] In summary, this embodiment of a visual recognition-based item sorting device involves directly pouring a fixed quantity of dates onto a feeding tray 200 during sorting. The vibration component 300 then causes the granules on the tray to quickly roll and disperse, falling into their respective positioning grooves 210. This creates a gap between adjacent granules, preventing the drawback of difficulty in identification and sorting due to accumulation. Subsequently, a laser emitter 800 vertically emits light spots, and the controller 900 calculates and establishes a coordinate system for the feeding tray 200 based on the two points and the captured image, ensuring each date has a corresponding coordinate point. Simultaneously, the controller 900 analyzes and judges the location and quantity of inferior dates based on detailed image information from multiple high-definition cameras 700. With the cooperation of the main telescopic electric cylinder 500, the X-axis adjusting electric cylinder 511, the Y-axis adjusting electric cylinder 512, and the clamp 600, defective granules are sorted and removed. This eliminates the need for individual identification and sorting of all winter dates, significantly improving sorting efficiency.

[0029] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in this invention can be combined with each other in any way. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A vision recognition based article sorting apparatus, characterized by, Including sorting table (100), the sorting table (100) upper central part is provided with a material placing disc (200) and a vibration assembly (300) for driving the material placing disc (200) to vibrate, the top plate (400) is fixedly arranged above the sorting table (100), the main telescopic cylinder (500) is movably connected to the bottom of the top plate (400) in the central part, the main telescopic cylinder (500) has a piston rod (501) at the bottom output end, and a clamp (600) at the end of the piston rod (501), further comprising an angle adjusting assembly (510) for driving the main telescopic cylinder (500) to rotate circumferentially, the main telescopic cylinder (500) is fixedly provided with a mounting disc (700) on the outer periphery of the lower end, a plurality of high-definition cameras (710) are distributed on the bottom outer periphery of the mounting disc (700), and a general control unit, namely a controller (900) of the sorting device is further included.

2. The vision recognition based article sorting device according to claim 1, characterized in that: The top surface of the material placing disc (200) is provided with a plurality of closely adjacent positioning grooves (210), and the outer periphery of the top of the material placing disc (200) has a material blocking plate (220).

3. The vision recognition based item sorting device according to claim 1, wherein: The vibration assembly (300) includes a plurality of spring legs (310) arranged in a circle on the bottom of the material placing disc (200), and a vibration motor (320) mounted in the central part of the bottom of the material placing disc (200), and the lower end of the spring leg (310) is fixedly installed on the top of the sorting table (100).

4. The vision recognition based item sorting device according to claim 1, wherein: The angle adjusting assembly (510) includes an X-axis adjusting cylinder (511) installed on the bottom of the top plate (400) and connected to one side of the main telescopic cylinder (500), a Y-axis adjusting cylinder (512) installed on the bottom of the top plate (400) and connected to the other side of the main telescopic cylinder (500), and a universal shaft (513) arranged at the head and tail of the X-axis adjusting cylinder (511), the head and tail of the Y-axis adjusting cylinder (512) and the top end of the main telescopic cylinder (500).

5. The vision recognition based item sorting device according to claim 1, wherein: The bottom of the top plate (400) is further provided with at least two laser emitters (800) connected to the controller (900), and the light beams generated by the laser emitters (800) are perpendicular to the material placing disc (200).

6. The vision recognition based item sorting device according to claim 1, wherein: The top of the sorting table (100) further has a concave annular waste collecting groove (110) distributed along the outer diameter of the material placing disc (200).

7. The vision recognition based item sorting device according to claim 1, wherein: The central part of the sorting table (100) further has an avoidance groove (120) corresponding to the vibration motor (320).

8. The vision recognition based item sorting device according to claim 1, wherein: The top of the sorting table (100) further has a support frame (410) for fixedly installing the top plate (400), and the controller (900) is installed on one side of the support frame (410).

Citation Information

Patent Citations

  • Automatic sorting device based on visual identification

    CN210816317U