Adjustable visual dark box system and adjusting method for deep learning detection

Abstract

The application discloses an adjustable visual dark box system and adjusting method for deep learning detection, comprising a visual dark box body, a camera assembly arranged in the visual dark box body, a light source assembly arranged in the visual dark box body, a working area for placing a detection object, an adjusting component arranged outside the visual dark box body, and a transmission mechanism connecting the adjusting component and the camera assembly and / or the light source assembly. The application solves the problems of inconvenient adjustment of the camera and the light source in the existing visual dark box, the need for opening the box operation, insufficient repeated positioning capability, and the imaging environment being easily disturbed, adjusts the internal camera and the light source without damaging the overall light shielding environment of the dark box, and thus improves the image acquisition adaptation capability, imaging consistency and detection stability under different detection object conditions.

Description

Technical Field

[0001] This invention relates to visual inspection using deep learning, specifically an adjustable visual darkroom system and adjustment method for deep learning inspection. Background Technology

[0002] In deep learning-based visual detection scenarios, the quality of image acquisition directly affects the model training effect and the final detection accuracy.

[0003] In practical applications, different objects being tested vary in size, height, material, and surface reflectivity, thus requiring different camera positions, shooting angles, light source positions, and illumination methods.

[0004] Most existing visual dark boxes employ a fixed or semi-fixed structure. The camera and light source are typically installed inside the dark box. When adjustments to the internal imaging conditions are needed to suit the object being inspected, the dark box often needs to be opened, and adjustments made manually by pushing or pulling, disassembling, or temporarily fixing it in place. This presents several problems: 1. The internal space of the camera obscura is limited, making adjustments inconvenient; 2. The adjustment process requires opening the box, which can easily disrupt the original stable lighting environment inside the darkroom; 3. Manual adjustment has poor repeatability, which is not conducive to forming consistent image acquisition conditions; 4. Switching efficiency is low in scenarios involving multiple detection targets and multiple rounds of training and data acquisition; 5. It can easily lead to fluctuations in the quality of acquired images, affecting the data consistency and stability of deep learning detection models.

[0005] Therefore, a visual darkbox system that can adapt to deep learning detection scenarios and allows for easy adjustment of the internal camera and light source status is needed to improve image acquisition quality and equipment utilization efficiency. Summary of the Invention

[0006] To address the shortcomings of the existing technology, this invention provides an adjustable visual darkbox system and adjustment method for deep learning detection. This invention solves the problems of inconvenient camera and light source adjustment, the need for opening the box, insufficient repeatability, and susceptibility to interference in the imaging environment in existing visual darkboxes. Without destroying the overall light-blocking environment of the darkbox, the internal camera and light source are adjusted, thereby improving the image acquisition adaptability, imaging consistency, and detection stability under different detection object conditions.

[0007] To achieve the above technical objectives, the present invention adopts the following technical solution: an adjustable visual darkbox system for deep learning detection, comprising, The visual dark box itself; Camera components disposed inside the visual dark box body; A light source assembly disposed inside the main body of the visual dark box; The work area used to place the objects to be inspected; Adjustment components are disposed on the outside of the visual dark box body; A transmission mechanism connecting the adjustment component to the camera assembly and / or the light source assembly.

[0008] The bottom of the visual dark box body is provided with a first through hole for accommodating the camera assembly and / or the light source assembly, and the back of the visual dark box body is provided with a second through hole. A part of the transmission mechanism is located inside the visual dark box body for connecting with the camera assembly and / or the light source assembly.

[0009] The visual dark box body includes a main box, a front panel, and a back panel. The bottom of the main box is provided with the first through hole. The front panel is located on the front side of the main box and has an observation window. The back panel is fixed to the back of the main box and the second through hole is located on the back panel.

[0010] The transmission mechanism includes at least one gear rack pair, each gear rack pair is connected to a bracket, the gear end of the at least one gear rack pair is connected to the adjustment component, and the bracket is connected to the camera assembly or the light source assembly.

[0011] The transmission mechanism comprises a vertical plate, a rack, and at least one gear. One of the gears and the rack form a gear-rack pair. The gear shaft of the gear is equipped with a lifting frame. The lifting frame is connected to the vertical plate through a guide rail slider assembly. The bracket is fixedly connected to the lifting frame.

[0012] The gears include a camera gear and a light source gear. The lifting frames are a camera lifting frame and a light source lifting frame, respectively. The camera lifting frame is connected to the gear shaft of the camera gear, and the light source lifting frame is connected to the gear shaft of the light source gear. The brackets are a camera bracket and a light source bracket. The camera bracket is fixed to the camera lifting frame, and the light source bracket is fixed to the light source lifting frame. The adjustment components include a camera adjustment handwheel and a light source adjustment handwheel. The camera adjustment handwheel is fixedly connected to the camera gear, and the light source adjustment handwheel is fixedly connected to the camera gear.

[0013] The camera bracket includes a U-shaped frame and a reinforcing rod. Support plates are provided on both side plates of the U-shaped frame. The two ends of the reinforcing rod are respectively fixed to the support plates. One side plate of the U-shaped frame is fixed to the camera lifting frame, and the other side plate extends into the interior of the visual dark box body and is fixedly connected to the camera assembly. The light source bracket includes a mounting plate and a fixing plate that are fixedly connected to each other. The mounting plate has a third through hole for the camera assembly to pass through. The mounting plate extends into the interior of the visual dark box body and is fixedly connected to the light source assembly. The fixing plate is fixedly connected to the light source lifting frame.

[0014] It also includes a cover, the visual dark box body is fixed to the cover, the transmission mechanism is located inside the cover, the cover is provided with a fourth through hole for the camera bracket and the light source bracket to pass through and extend into the visual dark box body, the cover is also provided with a fifth through hole for the camera adjustment handwheel and the light source adjustment handwheel to pass through, so that the handwheel ends of the camera adjustment handwheel and the light source adjustment handwheel are located outside the cover.

[0015] It also includes a base, the bottom of the cover is fixed to the base, the transmission mechanism is fixed to the base, the base is set with the working area, and the camera assembly and the light source assembly are both located directly above the working area.

[0016] An adjustment method for an adjustable visual darkbox system for deep learning detection includes the following steps: Manually rotate the camera adjustment handwheel, which drives the camera gear to rotate. Under the meshing of the camera gear and rack, the camera gear moves up and down, which in turn drives the camera lifting frame and camera bracket to move up and down, thereby driving the camera components to move up and down. Manually rotate the light source adjustment handwheel, which drives the light source gear to rotate. Under the meshing of the light source gear and the rack, the light source gear moves up and down, which in turn drives the light source lifting frame and the light source bracket to move up and down, thereby driving the light source assembly to move up and down.

[0017] In summary, the present invention has achieved the following technical effects: It supports camera and light source adjustment outside the darkroom, eliminating the need for operators to frequently open the darkroom, greatly optimizing the workflow, improving operational efficiency, and avoiding mechanical wear or positioning errors that may be caused by repeated opening and closing. The adjustment mechanism adopts an indirect transmission design to ensure that external light does not seep into the dark box during the adjustment process, thus maintaining a sealed light-blocking environment to the greatest extent and ensuring the high stability and repeatability of the imaging light source conditions. Both the camera and light source components have independent degrees of freedom adjustment, allowing users to make precise matches based on the size, shape, and optical characteristics of different objects being detected, significantly expanding the system's detection range and application scenarios. A stable imaging environment and repeatable component poses ensure high consistency of images acquired at different times for the same type of sample. This provides a high-quality, low-fluctuation data foundation for subsequent deep learning model training, while also ensuring the standardization of input data in actual detection and improving the reliability of analysis results. Its convenient adjustability and strong adaptability can not only meet the needs of high-speed industrial online testing, but also be used in various scenarios such as experimental research and development and teaching demonstrations that require frequent sample changes, making it highly practical. The present invention has strong functionality, stability and operability, and solves the problems of inconvenient adjustment, large environmental impact and poor light blocking effect in traditional methods. It has broad market potential in the field of precision vision inspection. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of an adjustable visual darkbox system for deep learning detection. Figure 2 yes Figure 1 A schematic diagram of the hidden visual dark box body; Figure 3 This is a schematic diagram of the elevation angle of the visual dark box body; Figure 4 This is a top-view diagram of the main body of the visual dark box; Figure 5 This is a diagram showing the front panel after the door is opened; Figure 6 This is a schematic diagram of the opening of the main body of the visual dark box; Figure 7 yes Figure 6 Another perspective illustration; Figure 8 This is an exploded view of the main body of the visual dark box after it has been opened; Figure 9 yes Figure 8 Another perspective illustration; Figure 10 This is a diagram of a camera bracket; Figure 11 This is a schematic diagram of the light source support; Figure 12 This is a schematic diagram of the cover; Figure 13 This is a diagram showing the position after adjustment; Figure 14 This is a schematic diagram of the hidden visual dark box. Detailed Implementation

[0019] The present invention will be further described in detail below with reference to the accompanying drawings.

[0020] This specific embodiment is merely an explanation of the present invention and is not intended to limit the invention. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of the claims of the present invention.

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0025] Example: Figure 1 This is a schematic diagram of an adjustable visual darkbox system for deep learning detection. Figure 2 yes Figure 1 A schematic diagram of the hidden visual dark box body 2, including, Visual dark box body 2; used to provide a semi-enclosed visual acquisition space, with a working area 9 set below the dark box to meet the sample acquisition needs of deep learning detection; Camera assembly 5 is installed inside the visual dark box body 2; The light source assembly 6 is located inside the visual dark box body 2; it is used to provide the required lighting conditions and adjust its position according to the surface features of the object being detected in order to improve the lighting effect and image quality. Working area 9 is used to place the object to be tested 10; Adjustment component 7 is located on the outside of the visual dark box body 2; the operator can adjust the internal structure through this component, avoiding direct insertion into the dark box for operation; The transmission mechanism 4 connects the adjustment component 7 to the camera assembly 5 and / or the light source assembly 6.

[0026] In practical use, the operator places the object to be detected in the working area inside the dark box. Based on the object's height, size, surface reflectivity, or detection requirements, the operator adjusts the position of the internal camera and light source using the external adjustment components of the dark box. The adjustment action is transmitted to the inside of the dark box via a transmission mechanism, so that the camera and light source components are in a state suitable for the current detection task. Subsequently, image acquisition is completed, and the acquired images are used for deep learning model training, inference detection, or experimental teaching.

[0027] This invention places the camera and light source components inside the visual dark chamber, while the adjustment mechanism is located outside the chamber. A transmission mechanism connects the adjustment mechanism to the camera and light source, allowing external control of the adjustment mechanism to adjust the position of the internal camera and light source. This achieves external control of the camera and light source height, making adjustment more convenient without disassembling the visual dark chamber or repeating positioning. This invention allows for adjustment of the internal camera and light source without disrupting the overall light-blocking environment of the dark chamber, thereby improving image acquisition adaptability, imaging consistency, and detection stability under different detection conditions.

[0028] Figure 3 This is a schematic diagram of the elevation angle of the main body 2 of the visual dark box. Figure 4This is a top-view schematic diagram of the visual dark box body 2. The bottom of the visual dark box body 2 is provided with a first through hole 21 for accommodating the camera assembly 5 and / or the light source assembly 6. The back of the visual dark box body 2 is provided with a second through hole 2033 for passing through a part of the transmission mechanism 4, so that a part of the transmission mechanism 4 passes through the second through hole 2033 and is located inside the visual dark box body 2 for connecting with the camera assembly 5 and / or the light source assembly 6.

[0029] The visual dark box body 2 forms a light-sealed cavity, with only two through holes at key functional locations. The first through hole is located at the bottom, providing necessary installation and adjustment space for the camera assembly 5 and the light source assembly 6, ensuring adjustable field of view and illumination angle. The second through hole is located at the rear, providing a through channel for the support of the transmission mechanism 4, ensuring smooth vertical adjustment of the entire camera-light source module. This invention ensures light leakage and interference prevention inside the dark box while precisely guaranteeing the adjustability and freedom of movement of key components, balancing sealing and functionality.

[0030] The visual dark box body 2 includes a main box 201, a front panel 202, and a back panel 203. The bottom of the main box 201 is provided with a first through hole 21. The front panel 202 is located on the front side of the main box 201 and is provided with an observation window 2021. The back panel 203 is fixed to the back of the main box 201 and a second through hole 2033 is provided on the back panel 203.

[0031] The observation window 2021 is made of transparent material, making it easy to observe the internal situation.

[0032] For ease of maintenance, the front panel 202 is designed as a door structure. Figure 5 This is a diagram showing the front panel 202 after the door is opened. Figure 6 This is a schematic diagram of the opening of the main body 2 of the visual dark box. Figure 7 yes Figure 6 Another perspective diagram, Figure 8 This is an exploded view of the main body of the visual dark box 2 after it is opened. Figure 9 yes Figure 8 Another perspective diagram.

[0033] Specifically, the main body 201 is a U-shaped structure consisting of two side panels 2011 and a top panel 2012, forming a structure with front and rear openings, left and right closures (the two side panels 2011 are the left and right sides), a top closure (top panel 2012), and a bottom opening (first through hole 21). The front opening is closed by the front panel 202, and the rear opening is closed by the back panel 203, thus forming a structure with only the bottom (first through hole 21) opening, forming a dark box structure.

[0034] More specifically, such as Figure 8As shown, the edges of the top plate 2012 extend beyond a certain distance, which is more conducive to light blocking. Light-blocking strips 2017 and 2018 are provided along the edges of the front opening for contact with the front panel 202, further enhancing light blocking. Similarly, the rear opening also has a light-blocking strip, not shown. Both side plates 2011 are equipped with second connecting posts 2016. The second fastening holes 2037 of the rear back plate 203 are bolted to the second connecting posts 2016, thus forming a secure connection between the rear back plate 203 and the main housing 201. A reinforcing strip 2019 is also provided at the bottom of the front opening, serving both as reinforcement and a light-blocking strip. The two ends of the reinforcing strip 2019 are bolted to the light-blocking strip. The first side panel 2011 is equipped with a first connecting post 2014, and the front panel 202 is equipped with a first fastening hole 2022. The first fastening hole 2022 is bolted to the first connecting post 2014, which can be fastened when the front panel 202 is closed. Next to the first connecting post 2014, a magnetic post 2015 is also provided. The front panel 202 is equipped with a magnet 204. When the front panel 202 is closed, it can be magnetically attracted, allowing the front panel 202 to be pre-closed, before being bolted to the first connecting post 2014 using the first fastening hole 2022. The magnetic post 2015 is made of the same material as the magnet 204. The second side panel 2011 is fitted with a hinge 2013, and the hinge hole 2026 of the front panel 202 connects to the hinge 2013, making the front panel 202 and the main body 201 a hinged door type.

[0035] The front panel 202 includes a middle plate 2024, a first side panel 2022, and a second side panel 2023, which together form an integral U-shaped structure. The second side panel 2023 and the first side panel 2022 cover the outer sides of the two side panels 2011 of the main body 201. The middle plate 2024 covers the front opening of the main body 201 and contacts the light-shielding strip. The second side panel 2023 has hinge holes 2026 for connection with hinges 2013, and a magnet 204 is fixed to the inner side of the first side panel 2022. Figure 7 As shown), the side panel 2022 is also provided with a fixing hole 2025, which passes through the magnet 204 and can communicate with the first connecting post 2014 to facilitate bolt installation.

[0036] The back panel 203 includes a first panel 2034, a second panel 2035, and a third panel 2036, which form an integral U-shaped structure. The first panel 2034 covers the rear opening of the main housing 201 and contacts the light-shielding strip. The second panel 2035 and the third panel 2036 cover the outer sides of the two side panels 2011 of the main housing 201. Both the second panel 2035 and the third panel 2036 are provided with second fastening holes 2037, which are connected to the second connecting posts 2016 of the two side panels 2011. A second through hole 2033 is provided in the first panel 2034. A bottom plate 2031 is provided at the bottom of the back panel 203, and its three sides are connected to the first panel 2034, the second panel 2035, and the third panel 2036 respectively. The edge of the bottom plate 2031 serves as the edge of the first through hole 21, and it is provided with an arc-shaped structure 2032 to adapt to the circular light source structure. The first panel 2034 is also provided with holes for fixed connection with the cover 1.

[0037] The transmission mechanism 4 includes at least one gear rack pair, each gear rack pair is connected to a bracket, the gear end of at least one gear rack pair is connected to an adjustment component 7, and the bracket is connected to the camera assembly 5 or the light source assembly 6.

[0038] In practical applications, one of the following three configuration methods can be adopted according to specific testing requirements: Fixed camera, adjustable light source configuration: The position of the camera assembly 5 remains fixed, while the position of the light source assembly 6 is adjusted via the transmission mechanism 4. This configuration is suitable for scenarios requiring a stable shooting angle and only needing to adjust the lighting conditions. In this configuration, the camera assembly 5 is pre-fixed inside the visual dark box body 2, and the light source assembly 6 is connected to the transmission mechanism 4.

[0039] Fixed light source, adjustable camera configuration: The light source assembly 6 is fixedly installed, while the camera assembly 5 can be raised, lowered, and its angle adjusted via the transmission mechanism 4. This configuration is suitable for detection tasks where the lighting environment is stable but the shooting position or angle needs to be flexibly changed. In this configuration, the light source assembly 6 is pre-fixed inside the visual dark box body 2, and the camera assembly 5 is connected to the transmission mechanism 4.

[0040] Both the camera and the light source are adjustable: both the camera assembly 5 and the light source assembly 6 can be adjusted independently via the transmission mechanism 4. This configuration offers maximum flexibility, allowing for the coordinated adjustment of their relative positions to accommodate various inspection objects 10 with different sizes, shapes, and surface characteristics, significantly enhancing the system's versatility and inspection performance optimization capabilities.

[0041] To fully illustrate the adjustment flexibility and technical advantages of this invention, this embodiment will be specifically described with both the camera and the light source being adjustable. This not only ensures that the system can flexibly respond to various detection needs, but also reflects significant progress in integration and adaptability.

[0042] like Figure 2As shown, the transmission mechanism 4 adopts a vertical plate 401, a rack 402 (with teeth), and at least one gear. The gear and the rack 402 form a gear rack pair. The gear shaft of the gear is equipped with a lifting frame. The lifting frame is connected to the vertical plate 401 through a guide rail slider assembly. The bracket is fixedly connected to the lifting frame.

[0043] Furthermore, the gears include a camera gear (not shown) and a light source gear (not shown). The lifting frames are a camera lifting frame 403 and a light source lifting frame 404. The camera lifting frame 403 is connected to the gear shaft of the camera gear, and the light source lifting frame 404 is connected to the gear shaft of the light source gear. The brackets are a camera bracket 405 and a light source bracket 406. The camera bracket 405 is fixed to the camera lifting frame 403, and the light source bracket 406 is fixed to the light source lifting frame 404. The adjustment component 7 includes a camera adjustment handwheel 71 and a light source adjustment handwheel 72. The camera adjustment handwheel 71 is fixedly connected to the camera gear, and the light source adjustment handwheel 72 is fixedly connected to the camera gear.

[0044] This invention features two gears mounted on rack 402. These gears are located inside camera lifting frame 403 and light source lifting frame 404 (not shown in the diagram). Specifically, camera lifting frame 403 and light source lifting frame 404 have identical structures. Taking camera lifting frame 403 as an example, it is a U-shaped frame. The gear shaft of the camera gear is mounted on two side plates of the U-shaped frame via two bearings, creating a rotatable connection between the gear shaft and the U-shaped frame. The U-shaped frame is slidably connected to the upright plate 401, allowing for the use of guide rails and sliders. Camera adjustment handwheel 701 is fixedly connected to the gear shaft. When camera adjustment handwheel 701 is rotated, the camera gear rotates accordingly, meshing with rack 402 while rack 402 remains stationary. This causes the camera gear to rise and fall, driving camera lifting frame 403 to rise and fall, which in turn drives camera bracket 405 to rise and fall, ultimately driving camera assembly 5 to rise and fall. The principle of light source lifting frame 404 is the same and will not be described further here.

[0045] Figure 10 This is a schematic diagram of camera bracket 405. Camera bracket 405 includes U-shaped frame 4051 and reinforcing rod 4052. Both side plates of U-shaped frame 4051 are provided with support plates 4053. The two ends of reinforcing rod 4052 are respectively fixed to the support plates 4053. One side plate of U-shaped frame 4051 is fixed to camera lifting frame 403, and the other side plate extends into the interior of visual dark box body 2 and is fixedly connected to camera assembly 5.

[0046] The bracket of the present invention can connect the camera to the transmission mechanism 4, so that when the transmission mechanism 4 is controlled to lift and lower, it can drive the camera to lift and lower. One end of the bracket is located inside the visual dark box body 2 and the other end is located outside the visual dark box body 2, thereby transmitting the external drive to the inside, realizing external adjustment to drive the internal movement.

[0047] Figure 11 This is a schematic diagram of the light source bracket 406. The light source bracket 406 includes a mounting plate 4061 and a fixing plate 4062 that are fixedly connected to each other. The mounting plate 4061 is provided with a third through hole 4063 for the camera assembly 5 to pass through. The mounting plate 4061 extends into the interior of the visual dark box body 2 and is fixedly connected to the light source assembly 6. The fixing plate 4062 is fixedly connected to the light source lifting frame 404.

[0048] The light source bracket 406 and camera bracket 405 are designed based on the same principle. While maintaining structural coordination, they also provide sufficient room for movement for the camera assembly 5 to accommodate its position adjustment needs. The camera assembly 6 generally adopts a ring structure, with the camera assembly 5 mounted in the central area within the ring. This surrounding layout not only makes the camera's field of view more focused and the light utilization more efficient, but also reduces external obstructions, thereby obtaining a clearer and more uniformly illuminated imaging effect.

[0049] like Figure 1 As shown, it also includes a cover 1, a visual dark box body 2 fixed to the cover 1, and a transmission mechanism 4 located inside the cover 1. Figure 12 This is a schematic diagram of the enclosure 1. The enclosure 1 is provided with a fourth through hole 101 through which the camera bracket 405 and the light source bracket 406 extend into the visual dark box body 2. The enclosure 1 is also provided with a fifth through hole 102 through which the camera adjustment handwheel 71 and the light source adjustment handwheel 72 extend, so that the handwheel ends of the camera adjustment handwheel 71 and the light source adjustment handwheel 72 are located outside the enclosure 1.

[0050] The cover 1 has a bottom opening structure. A fourth through hole 101 is provided on the front to allow for the two supports to move. A fifth through hole 102 is provided on the side to lead the handwheel to the outside.

[0051] like Figure 1 As shown, it also includes a base 3, the bottom of the cover 1 is fixed to the base 3, the transmission mechanism 4 is fixed to the base 3, the base 3 is provided with a working area 9, and the camera assembly 5 and the light source assembly 6 are both located directly above the working area 9.

[0052] The base 1 also houses the electrical modules required for the dark box, such as the power module. The electrical modules are common models, and their structure, principle, and connection methods are existing technologies, which will not be described in detail here.

[0053] In another embodiment, the present invention also provides an adjustment method for an adjustable visual darkbox system for deep learning detection, comprising the following steps: 1. Manually rotate the camera adjustment handwheel 71. The camera adjustment handwheel 71 drives the camera gear to rotate. Under the meshing of the camera gear and the rack 42, the camera gear makes a lifting motion, which drives the camera lifting frame 403 and the camera bracket 405 to make a lifting motion, thereby driving the camera assembly 5 to lift. 2. Manually rotate the light source adjustment handwheel 72. The light source adjustment handwheel 72 drives the light source gear to rotate. Under the meshing of the light source gear and the rack 42, the light source gear makes a lifting motion, which drives the light source lifting frame 404 and the light source bracket 406 to make a lifting motion, thereby driving the light source assembly 6 to lift.

[0054] The camera adjustment handwheel 71 and the light source adjustment handwheel 72 of the present invention can be controlled independently, and can independently adjust the camera or the light source, making it convenient to adjust the position of the camera and the light source according to the actual situation. Figure 13 This is a schematic diagram after the position is adjusted. In the diagram, the camera component 5 is adjusted upwards, changing its position relative to the light source component 6.

[0055] Figure 14 This is a schematic diagram of the hidden visual dark box body 2. The fifth through hole 102 of the cover 1 is equipped with a flexible baffle 103. During the lifting and lowering process of the two handwheels, the flexible baffle 103 will not affect the movement of the handwheels, and at the same time can play the role of blocking light and preventing dust.

[0056] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall fall within the scope of the technical solution of the present invention.

Claims

1. An adjustable visual darkbox system for deep learning detection, characterized in that: include, Visual dark box body (2); Camera assembly (5) disposed inside the visual dark box body (2); A light source assembly (6) is disposed inside the visual dark box body (2); The work area (9) is used to place the test object (10); Adjustment component (7) disposed outside the main body (2) of the visual dark box; A transmission mechanism (4) connecting the adjustment component (7) to the camera assembly (5) and / or the light source assembly (6).

2. The adjustable visual darkbox system for deep learning detection according to claim 1, characterized in that: The bottom of the visual dark box body (2) is provided with a first through hole (21) for accommodating the camera assembly (5) and / or the light source assembly (6). The back of the visual dark box body (2) is provided with a second through hole (2033). A part of the transmission mechanism (4) is located inside the visual dark box body (2) for connecting with the camera assembly (5) and / or the light source assembly (6).

3. The adjustable visual darkbox system for deep learning detection according to claim 2, characterized in that: The visual dark box body (2) includes a main box (201), a front panel (202), and a back panel (203). The bottom of the main box (201) is provided with the first through hole (21). The front panel (202) is located on the front side of the main box (201). The back panel (203) is fixed to the back of the main box (201). The second through hole (2033) is located on the back panel (203).

4. The adjustable visual darkbox system for deep learning detection according to claim 1, characterized in that: The transmission mechanism (4) includes at least one gear rack pair, each gear rack pair is connected to a bracket, the gear end of the at least one gear rack pair is connected to the adjustment component (7), and the bracket is connected to the camera assembly (5) or the light source assembly (6).

5. An adjustable visual darkbox system for deep learning detection according to claim 4, characterized in that: The transmission mechanism (4) consists of a vertical plate (401), a rack (402), and at least one gear. The gear and the rack (402) form a gear rack pair. The gear shaft of the gear is equipped with a lifting frame. The lifting frame is connected to the vertical plate (401) through a guide rail slider assembly. The bracket is fixedly connected to the lifting frame.

6. The adjustable visual darkbox system for deep learning detection according to claim 5, characterized in that: The gears include a camera gear and a light source gear. The lifting frames are a camera lifting frame (403) and a light source lifting frame (404). The camera lifting frame (403) is connected to the gear shaft of the camera gear, and the light source lifting frame (404) is connected to the gear shaft of the light source gear. The brackets are a camera bracket (405) and a light source bracket (406). The camera bracket (405) is fixed to the camera lifting frame (403), and the light source bracket (406) is fixed to the light source lifting frame (404). The adjustment component (7) includes a camera adjustment handwheel (71) and a light source adjustment handwheel (72). The camera adjustment handwheel (71) is fixedly connected to the camera gear, and the light source adjustment handwheel (72) is fixedly connected to the camera gear.

7. The adjustable visual darkbox system for deep learning detection according to claim 6, characterized in that: The camera bracket (405) includes a U-shaped frame (4051) and a reinforcing rod (4052). The two side plates of the U-shaped frame (4051) are provided with support plates (4053). The two ends of the reinforcing rod (4052) are respectively fixed to the support plates (4053). One side plate of the U-shaped frame (4051) is fixed to the camera lifting frame (403), and the other side plate extends into the interior of the visual dark box body (2) and is fixedly connected to the camera assembly (5). The light source bracket (406) includes a mounting plate (4061) and a fixing plate (4062) that are fixedly connected to each other. The mounting plate (4061) is provided with a third through hole (4063) for the camera assembly (5) to pass through. The mounting plate (4061) extends into the interior of the visual dark box body (2) and is fixedly connected to the light source assembly (6). The fixing plate (4062) is fixedly connected to the light source lifting frame (404).

8. An adjustable visual darkbox system for deep learning detection according to claim 6, characterized in that: It also includes a cover (1), the visual dark box body (2) is fixed to the cover (1), the transmission mechanism (4) is located inside the cover (1), the cover (1) is provided with a fourth through hole (101) for the camera bracket (405) and the light source bracket (406) to pass through and extend into the visual dark box body (2), the cover (1) is also provided with a fifth through hole (102) for the camera adjustment handwheel (71) and the light source adjustment handwheel (72) to pass through, so that the handwheel ends of the camera adjustment handwheel (71) and the light source adjustment handwheel (72) are located outside the cover (1).

9. An adjustable visual darkbox system for deep learning detection according to claim 8, characterized in that: It also includes a base (3), the bottom of the cover (1) is fixed to the base (3), the transmission mechanism (4) is fixed to the base (3), the base (3) sets the working area (9), and the camera assembly (5) and the light source assembly (6) are both located directly above the working area (9).

10. An adjustment method for an adjustable visual darkbox system for deep learning detection, characterized in that: An adjustable visual darkbox system for deep learning detection, as described in any one of claims 1-9, includes the following steps: Manually rotate the camera adjustment handwheel (71). The camera adjustment handwheel (71) drives the camera gear to rotate. Under the meshing of the camera gear and the rack (42), the camera gear makes a lifting motion, which drives the camera lifting frame (403) and the camera bracket (405) to make a lifting motion, thereby driving the camera assembly (5) to lift. Manually rotate the light source adjustment handwheel (72). The light source adjustment handwheel (72) drives the light source gear to rotate. Under the meshing of the light source gear and the rack (42), the light source gear makes a lifting motion, which drives the light source lifting frame (404) and the light source bracket (406) to make a lifting motion, thereby driving the light source assembly (6) to lift.

Images