Multifunctional experiment box based on computer vision

Abstract

The utility model provides a multifunctional experiment box based on computer vision, which comprises an aluminum box body, and the interior of the aluminum box body is connected with a monocular camera through a camera support rod and is connected with a binocular camera through a two-dimensional camera holder. The camera support rod and the two-dimensional camera holder are sleeved with an acrylic interlayer which is penetrated by the acrylic interlayer and arranged in the aluminum box body, and one side of the acrylic interlayer is connected with a two-finger clamping jaw through a five-degree-of-freedom mechanical arm connected with the interior of the aluminum box body. A jetson orin nano calculation board connected with the monocular camera, the binocular camera and the five-degree-of-freedom mechanical arm is arranged on one side of the interior of the aluminum box body, obtained visual images are processed through the jetson orin nano calculation board through visual systems of the monocular camera 2 and the binocular camera 11 arranged in the aluminum box body, control commands are issued to the mechanical arm and the two-finger clamping jaw, and the mechanical arm and the two-finger clamping jaw are controlled to work. Corresponding operation is completed; the camera support rod and the two-dimensional camera holder can enhance the stable operation of the camera.

Description

Technical Field

[0001] The utility model provides a multifunctional experiment box, belongs to the technical field of computer vision equipment, and particularly relates to a multifunctional experiment box based on computer vision. Background Art

[0002] A multifunctional computer vision experiment box integrates image processing, pattern recognition, and data analysis capabilities, designed to support a wide range of scientific research and engineering experiments. Its background technology encompasses advanced image acquisition techniques, deep learning algorithms, and high-performance computing hardware, enabling the real-time capture, analysis, and interpretation of complex visual data. This experiment box is commonly used for applications such as object recognition, motion tracking, and behavioral analysis, providing researchers and engineers with a flexible platform for exploring and developing new visual computing technologies and their applications across various fields.

[0003] There are many computer vision experiment kits available on the market for AI applications, but these lack versatility and focus on AI vision. They typically feature a monocular camera and an edge computing board, with some also adding robotic arms and other sensor equipment. These edge computing boards have relatively low computing power and are typically used for specific project deployments, making them less suitable for basic teaching and general development. Furthermore, with the rise of 3D vision technology, monocular cameras are clearly not sufficient for learning and research. Utility Model Content

[0004] Other AI-based computer vision experiment boxes currently on the market have problems with unstable foundations, insufficient depth, and poor scalability. Other AI-based computer vision experiment boxes currently on the market have problems with unstable foundations, insufficient depth, and poor scalability.

[0005] In order to solve the above technical problems, the utility model provides the following technical solutions: a multifunctional experimental box based on computer vision, comprising an aluminum box body, the interior of the aluminum box body is connected to a monocular camera through a camera bracket rod, and a binocular camera is connected through a two-dimensional camera gimbal, the outer sleeve of the camera bracket rod and the two-dimensional camera gimbal is provided with an acrylic interlayer that penetrates itself and is placed inside the aluminum box body, one side of the acrylic interlayer is connected to a two-finger gripper through a 5-degree-of-freedom robotic arm connected to the inside of the aluminum box body, and one side of the inside of the aluminum box body is provided with a jetsonorinnano computing board connected to the monocular camera, the binocular camera and the 5-degree-of-freedom robotic arm.

[0006] Preferably, the monocular camera is movably connected to the camera support rod through a monocular camera fastener, the camera support rod is provided with a plurality of support fasteners, and the end of the camera support rod away from the monocular camera is connected to a camera support fixing base placed under the acrylic interlayer.

[0007] Preferably: a power supply base plate electrically connected to the jetsonorinnano computing board is provided under the jetsonorinnano computing board, an edge computer base is provided between the power supply base plate and the jetsonorinnano computing board, and a high-definition display screen is provided on one side of the power supply base plate and electrically connected to the acrylic interlayer, and a display screen base is provided between the acrylic interlayer.

[0008] Preferably: a sorting tool corresponding to the two-finger gripper is provided above the acrylic partition, a robotic arm fixing base is provided below the two-finger gripper and is placed below the acrylic partition and connected to the 5-DOF robotic arm, and a keyboard is provided below the high-definition display screen and is placed below the acrylic partition.

[0009] Preferably: a sorting warehouse corresponding to the sorting props is provided on one side of the fixed base of the robotic arm, the sorting warehouse and the bottom layer of the aluminum box body are both provided with EVA foam lining, and the outer cover of the binocular camera is provided with a binocular camera housing connected to the two-dimensional camera gimbal.

[0010] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0011] The utility model solves the problem of meeting the requirements of traditional visual scenes and 3D visual scenes at the same time by providing a binocular camera and a monocular camera.

[0012] The utility model solves the problem of providing an industrial sorting scene by arranging a mechanical arm, a sorting warehouse and sorting props.

[0013] The utility model solves the problem of supporting intelligent security, identity authentication and automatic obstacle avoidance scenarios by combining a binocular camera with a two-dimensional pan-tilt platform.

[0014] This utility model solves the problem of insufficient computing power by setting up a high-performance Jetsonorinnano edge computing board.

[0015] The utility model solves the problem of supplying power to each component by arranging an independent power supply base plate.

[0016] The utility model solves the problem of storing accessories by arranging an EVA foam lining structure.

[0017] The utility model solves the problems of camera position fixation and high customization by arranging a square camera support rod and connecting it with a fastener.

[0018] The utility model solves the problem of unclear screen viewing by providing an 11.6-inch high-definition display screen with an inclined angle.

[0019] The utility model sets up a visual system of a monocular camera and a binocular camera placed inside an aluminum box. After the obtained visual images are processed by a Jetson Orin Nano computing board, control commands are issued to the robotic arm and the two-finger gripper to complete the corresponding operations. The camera support rod and the two-dimensional camera pan-tilt platform can enhance the stable operation of the camera.

[0020] Other advantages, objectives and features of the present invention will be described in part in the following description and will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a front view of a multifunctional experimental box based on computer vision in the present utility model;

[0022] Figure 2 This is a three-dimensional schematic diagram of another perspective of a multifunctional experimental box based on computer vision in the present utility model;

[0023] Figure 3 This is a three-dimensional schematic diagram of the lower layer of the acrylic partition of a multifunctional experimental box based on computer vision in the present utility model.

[0024] As shown in the figure:

[0025] 1. Aluminum box; 2. Monocular camera; 3. Monocular camera fasteners; 4. Camera bracket rod; 5. Bracket fasteners; 6. HD display; 7. Power supply base; 8. Jetsonorinnano computing board; 9. Edge computing board base; 10. Binocular camera housing; 11. Binocular camera; 12. 2D camera gimbal; 13. Camera bracket fixing base; 14. Acrylic interlayer; 15. Camera bracket fixing base; 16. Robotic arm fixing base; 17. 5-DOF robotic arm; 18. Two-finger gripper; 19. Sorting bin; 20. EVA foam lining; 21. Display base; 22. Keyboard

[0026] 141. Separate props. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0028] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are for illustrative purposes only and do not represent the only implementation method.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of this invention are for the purpose of describing specific embodiments only and are not intended to limit this invention; the term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0030] like Figure 1 and Figure 2 As shown, a multifunctional experimental box based on computer vision includes an aluminum box body 1, a monocular camera 2 connected to the inside of the aluminum box body 1 through a camera bracket rod 4, and a binocular camera 11 connected through a two-dimensional camera gimbal 12, the monocular camera 2 is movably connected to the camera bracket rod 4 through a monocular camera fastener 3, and a plurality of bracket fasteners 5 are provided on the camera bracket rod 4, and the end of the camera bracket rod 4 away from the monocular camera 2 is connected to a camera bracket fixing base 15 placed under an acrylic interlayer 14, and the outer sleeve of the camera bracket rod 4 and the two-dimensional camera gimbal 12 is provided with an acrylic interlayer 14 that is penetrated by itself and placed inside the aluminum box body 1, and one side of the acrylic interlayer 14 is connected to a two-finger gripper 18 through a 5-DOF robotic arm 17 connected to the inside of the aluminum box body 1, and one side of the inside of the aluminum box body 1 is provided with a jetsonorinnano computing board 8 connected to the monocular camera 2, the binocular camera 11 and the 5-DOF robotic arm 17.

[0031] In this implementation scheme, the utility model is equipped with both binocular cameras and monocular cameras, which can meet the needs of traditional vision and 3D vision scenarios, and is suitable for various purposes such as teaching, research and development; it uses the Jetsonorinnano edge computing board, which has powerful computing power and comes with a CUDA environment, and can directly run advanced deep learning projects. The monocular camera is combined with a robotic arm and a sorting warehouse to support industrial sorting scenarios; the binocular camera is combined with a two-dimensional gimbal to support scenarios such as intelligent security, identity authentication and automatic obstacle avoidance. The structural design of the utility model is optimized, and the independent power supply base plate supplies power to each device. The display base has a tilt angle for easy viewing. The camera bracket adopts a plug-in design, which is easy to install and fixed in position. The bottom is lined with EVA foam to store various accessories; the camera bracket height is customizable, and the robotic arm can be disassembled for storage, providing flexibility for different experimental needs. In general, this experimental box has significant improvements in function, performance and design, and can better meet the diverse needs of the field of AI computer vision.

[0032] like Figure 2 and Figure 3 As shown, a power supply base plate 7 electrically connected to the jetsonorinnano computing board 8 is provided below the jetsonorinnano computing board 8, an edge computer base 9 sleeved on the outside of the jetsonorinnano computing board 8 is provided between the power supply base plate 7 and the jetsonorinnano computing board 8, a high-definition display screen 6 electrically connected to the high-definition display screen 6 is provided on one side of the power supply base plate 7, and a display screen base 21 is provided between the high-definition display screen 6 and the acrylic interlayer 14, and a two-finger clamp 1 is provided above the acrylic interlayer 14. 8 corresponding to the sorting props 141, the two-finger gripper 18 is provided below the robotic arm fixed base 16 placed below the acrylic partition 14 and connected to the 5-DOF robotic arm 17, a keyboard 22 is provided below the high-definition display screen 6 and placed below the acrylic partition 14, one side of the robotic arm fixed base 16 is provided with a sorting warehouse 19 corresponding to the sorting props 141, the sorting warehouse 19 and the bottom layer of the aluminum box body 1 are both provided with an EVA foam lining 20, and the outer cover of the binocular camera 11 is provided with a binocular camera housing 10 connected to the two-dimensional camera gimbal 12.

[0033] In this implementation scheme, core components such as the Jetsonorinnano computing board, power supply baseboard, and high-definition display are compactly integrated to form a fully functional computing and display system; the power supply baseboard provides power for multiple devices, simplifies wiring, and improves reliability; the high-definition display is installed through the display base with a certain tilt angle for easy viewing; the combination of monocular camera, robotic arm, sorting props and sorting warehouse supports industrial sorting scenarios; the combination of binocular camera and two-dimensional pan-tilt head supports object tracking and 3D vision applications; the acrylic partition divides the space into two layers, the upper layer is the operation area, and the lower layer is the storage area; the EVA foam lining can store various accessories and make full use of the internal space of the box; such as the robotic arm fixed base, camera bracket, etc., are easy to install, disassemble and fix in position; working areas for different scenarios are reserved, which can be flexibly configured according to needs.

[0034] When using:

[0035] 1. Open the aluminum box and take out the required components.

[0036] 2. Install the monocular camera:

[0037] Insert the camera stand rod into the camera stand fixing base

[0038] Use the monocular camera fastener to fix the monocular camera on the top of the bracket

[0039] Camera height can be adjusted via bracket fasteners

[0040] 3. Install the binocular camera:

[0041] Install the binocular camera into the binocular camera housing

[0042] Fix the housing on the 2D camera gimbal

[0043] 4. Install the robotic arm:

[0044] Fix the 5-DOF robotic arm on the robotic arm fixed base

[0045] Install a two-finger gripper at the end of the robotic arm

[0046] 5. Arrange the work area:

[0047] Place sorting tools on top of the acrylic divider

[0048] Place sorting bins at appropriate locations

[0049] 6. Connect the power and data cables:

[0050] Connect each device to the power supply baseboard

[0051] Connect the device to the Jetsonorinnano computing board

[0052] 7. Start the system:

[0053] Turn on the power and start the jetsonorinnano computing board

[0054] Operate the system via a high-definition display and keyboard

[0055] 8. Start the experiment:

[0056] Can conduct various experiments such as industrial sorting, object tracking, 3D vision, etc.

[0057] Adjust the camera position, robotic arm posture, etc. as needed

[0058] 9. End of the experiment:

[0059] Disassemble and store each component into the corresponding slot of the EVA foam lining

[0060] Turn off the power and close the cover

[0061] This experimental box is flexibly designed and can be configured according to different experimental needs, making it convenient for users to carry out various computer vision-related teaching, research and development work.

[0062] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this technology may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the claims.

Claims

1. A multifunctional experimental box based on computer vision, comprising an aluminum box body (1), characterized in that: The interior of the aluminum box body (1) is connected to a monocular camera (2) via a camera support rod (4) and to a binocular camera (11) via a two-dimensional camera platform (12); the camera support rod (4) and the two-dimensional camera platform (12) are externally sleeved with an acrylic interlayer (14) which penetrates the acrylic interlayer and is placed inside the aluminum box body (1); one side of the acrylic interlayer (14) is connected to a two-finger gripper (18) via a 5-degree-of-freedom robotic arm (17) connected to the interior of the aluminum box body (1); and one side of the interior of the aluminum box body (1) is provided with a Jetson Orin Nano computing board (8) which is connected to the monocular camera (2), the binocular camera (11) and the 5-degree-of-freedom robotic arm (17).

2. The multifunctional experiment box based on computer vision according to claim 1, characterized in that: The monocular camera (2) is movably connected to a camera support rod (4) via a monocular camera fastener (3); a plurality of support fasteners (5) are provided on the camera support rod (4); and an end of the camera support rod (4) away from the monocular camera (2) is connected to a camera support fixing base (15) placed below an acrylic interlayer (14).

3. The multifunctional experiment box based on computer vision according to claim 1, characterized in that: A power supply base plate (7) electrically connected to the Jetson Orin Nano computing board (8) is provided below the Jetson Orin Nano computing board (8); an edge computer base (9) sleeved on the outside of the Jetson Orin Nano computing board (8) is provided between the power supply base plate (7) and the Jetson Orin Nano computing board (8); a high-definition display screen (6) placed above an acrylic interlayer (14) and electrically connected to the Jetson Orin Nano computing board (8) is provided on one side of the power supply base plate (7); and a display screen base (21) is provided between the high-definition display screen (6) and the acrylic interlayer (14).

4. The multifunctional experiment box based on computer vision according to claim 3, characterized in that: A sorting tool (141) corresponding to the two-finger gripper (18) is provided above the acrylic interlayer (14); a mechanical arm fixing base (16) disposed below the acrylic interlayer (14) and connected to the 5-DOF mechanical arm (17) is provided below the two-finger gripper (18); and a keyboard (22) disposed below the acrylic interlayer (14) is provided below the high-definition display screen (6).

5. The multifunctional experiment box based on computer vision according to claim 4, characterized in that: A sorting warehouse (19) corresponding to the sorting props (141) is provided on one side of the fixed base (16) of the robotic arm. The bottom layers of the sorting warehouse (19) and the aluminum box body (1) are both provided with an EVA foam lining (20). The outer surface of the binocular camera (11) is provided with a binocular camera housing (10) connected to a two-dimensional camera platform (12).

Images