A visual detection module, a robot head device and a robot

Abstract

The utility model discloses a kind of visual inspection module, robot head device and robot, by setting multiple field of view angle and resolution different image acquisition piece, first wide-range image information acquisition is carried out, the position of target object is determined, target object is moved to the field of view range of higher resolution image acquisition piece later, image acquisition and positioning are carried out, higher resolution image information acquisition is carried out one by one, adjusting target object into field of view range is realized, and the image information acquisition of high definition to target object is realized.The utility model provides a kind of visual inspection module, comprising: multiple image acquisition pieces, the field of view angle and resolution of different image acquisition pieces are different, and image acquisition piece is used to obtain the image information of target object;Support, image acquisition piece is set on support, to make the relative position of different image acquisition pieces fixed.The utility model is mainly used for the optical detection of to-be-tested object.

Description

Technical Field

[0001] This utility model relates to the field of machine vision technology, and in particular to a visual inspection module, a robot head device, and a robot. Background Technology

[0002] Traditional automated inspection equipment based on machine vision is costly and lacks versatility. In recent years, the fields of machine vision and artificial intelligence have developed rapidly, exhibiting characteristics of multidisciplinary integration and hardware-software synergy. Humanoid robots, as the most promising products, are flourishing and being applied in various fields such as material handling and equipment maintenance. Machine vision provides robots with environmental perception and target recognition functions, acting as their "eyes," enabling robots to replace humans in completing some tasks in harsh environments.

[0003] In existing humanoid robots, visual inspection typically uses a single image acquisition device. If the field of view of the image acquisition device is large, the resolution will decrease, making it difficult to acquire clear image information and resulting in inaccurate detection results. Conversely, if the resolution is high, the field of view will be small, requiring the arm to move multiple times to find a position within the robot's visual inspection range, thus reducing detection efficiency. Utility Model Content

[0004] In view of this, the present invention provides a visual inspection module, a robot head device, and a robot, mainly to solve the problem that existing robot visual inspection is difficult to balance efficiency and accuracy.

[0005] To achieve the above objectives, this utility model mainly provides the following technical solutions: On the one hand, this utility model provides a visual inspection module for robots, comprising: Multiple image acquisition units, each with different field of view and resolution, are used to acquire image information of the target object; The support (105) is used to fix the relative positions of different image acquisition components.

[0006] Among them, the optical axes of different image acquisition devices are parallel, the fields of view of different image acquisition devices have overlapping areas, and the optimal shooting distance of different image acquisition devices is the same.

[0007] Among them, multiple image acquisition components include a wide-angle camera lens module (101), a standard camera lens module (102), and a telephoto camera lens module (103). The field of view of the wide-angle camera lens module (101) is greater than that of the standard camera lens module (102) and the telephoto camera lens module (103); The resolution of the telephoto camera lens module (103) is greater than that of the wide-angle camera lens module (101) and the standard camera lens module (102).

[0008] The visual inspection module (100) also includes a fill light (104) which is used to provide ambient light.

[0009] Among them, the fill light (104) is a ring light, and the fill light (104) surrounds the periphery of multiple image acquisition components.

[0010] The bracket (105) includes a bottom wall and a side wall, the side wall surrounds the bottom wall and extends to one side of the bottom wall, and the image acquisition component is disposed on the bottom wall; The inner periphery of the sidewall includes an annular inclined surface extending toward the center of the area enclosed by the sidewall and simultaneously toward the bottom wall, and the supplementary light (104) is disposed on the annular inclined surface.

[0011] The bracket (105) includes a bottom wall and a side wall, the side wall surrounds the bottom wall and extends to one side of the bottom wall, and the image acquisition component is disposed on the bottom wall; The device also includes a face protector (106), which is made of transparent material and covers one side of the sidewall opposite to the bottom wall.

[0012] On the other hand, the present invention also provides a robot head device, including a vision detection module (100) as described above. Depth camera (200), the depth camera (200) is used to detect and acquire depth information; The support component (300), the visual inspection module (100) and the depth camera (200) are all connected to the support component (300).

[0013] The device also includes: An interactive element (400) is connected to a support component (300) and is used to receive sound signals and generate feedback signals.

[0014] The support assembly (300) includes a fastener (301) and a neck connection mechanism (302). Both the visual inspection module (100) and the depth camera (200) are connected to the fixture (301); The neck connection mechanism (302) includes a first adjustment component and a second adjustment component, the first adjustment component being connected to the second adjustment component, and the second adjustment component being connected to the fastener (301); The first adjustment component is used to drive the second adjustment component and the fixing member (301) to rotate around the rotation axis, and the second adjustment component is used to drive the fixing member (301) to swing relative to the swing axis.

[0015] The device also includes a controller (600), which is connected to the support assembly (300); The device also includes a heat dissipation component (500), which is directly or indirectly connected to the controller (600) for dissipating heat from the controller (600).

[0016] The heat dissipation component (500) includes a fan and / or multiple heat dissipation fins arranged in parallel.

[0017] On the other hand, this utility model discloses a robot, including a vision detection module (100) as described in any of the above, and a main body, a support (105) connected to the main body, the main body including at least an arm for picking up and moving target objects; Alternatively, it may include a robot head assembly as described above, and a body, a support assembly (300) connected to the body, the body including at least an arm for picking up and moving a target object.

[0018] This invention proposes a visual inspection module, a robot head device, and a robot. By setting up multiple image acquisition components with different fields of view and resolutions, it first acquires a wide range of image information using the large field-of-view component. Then, it determines the position of the target object based on the image information. Next, it moves the target object into the field of view of the higher-resolution image acquisition component, continuing image acquisition and positioning. This process of acquiring higher-resolution image information one by one allows for rapid adjustment of the target object into the field of view of the high-resolution image acquisition component, achieving high-definition image information acquisition and improving the accuracy and efficiency of detection. Furthermore, by treating the visual inspection module as a whole and fixing the relative positions of the image acquisition components with a bracket, the relative relationship of the field of view positions of the image acquisition components is unique, resulting in high accuracy. Compared to installing individual image acquisition components onto the robot, this invention eliminates the need for user calibration. Attached Figure Description

[0019] Figure 1 A schematic diagram of a robot head device from a first-view perspective, provided as an embodiment of this utility model; Figure 2 A schematic diagram of a robot head device provided in a second-view perspective for an embodiment of this utility model; Figure 3 A schematic diagram of a robot head device from a third-person perspective, provided as an embodiment of this utility model; The components include a visual inspection module 100, a bracket 105, a wide-angle camera lens module 101, a standard camera lens module 102, a telephoto camera lens module 103, a fill light 104, a face protection component 106, a depth camera 200, a support component 300, a fixing component 301, a neck connection mechanism 302, an interaction component 400, a heat dissipation component 500, and a controller 600. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by this utility model in order to achieve the intended utility model purpose, the following detailed description of the specific implementation method, structure, features and effects of the light source proposed according to this utility model is provided in conjunction with the accompanying drawings and preferred embodiments.

[0021] On one hand, this utility model provides a visual inspection module for a robot. The visual inspection module can be used in the robot's head, and the robot also includes a main body, which includes a robotic arm capable of grasping and moving target objects. The visual inspection module is used to acquire images of target objects, aiming to quickly align with the target object and capture clear images. However, in the initial stage of detection, the position of the target object is unknown, or rather, cannot be accurately estimated. Furthermore, when the robotic arm moves the target object, mechanical errors may cause errors in the target object's position. This utility model aims to quickly move the target object into the field of view of a high-resolution image acquisition device and acquire high-resolution image information.

[0022] Specifically, such as Figure 1-3 As shown, the visual inspection module 100 includes: Multiple image acquisition units, each with different field of view and resolution, are used to acquire image information of the target object; The image acquisition components are mounted on the bracket 105 to fix the relative positions of different image acquisition components.

[0023] The bracket 105 is used to connect image acquisition components to support and position them. Specifically, the bracket 105 may have multiple mounting holes, and the image acquisition components can be inserted into these holes. After connection, the angle of the optical axis of the image acquisition components and the relative positions between the multiple image acquisition components will be fixed. The structure of the mounting holes is adapted to the image acquisition components to ensure easy installation and prevent positional movement. The structures of the mounting holes can be the same or different. For example, when different image acquisition components have different external contour structures, the shapes of the mounting holes are set for different image acquisition components. In some embodiments, the bracket 105 includes a bottom wall, and the mounting holes are holes opened in the thickness direction of the bottom wall of the bracket 105, penetrating both opposite sides of the bottom wall in the thickness direction.

[0024] The image acquisition component can be any device used to acquire images, such as a camera. The camera mainly includes a lens and a circuit board. The lens includes an outer barrel and at least one optical lens located within the outer barrel. The optical axis of the image acquisition component refers to the optical axis of the optical lens, and the optical axis is typically perpendicular to the circuit board. The image acquisition component can be mounted via a hole on one side of the lens, and the circuit board is located on the back of the bottom wall of the bracket 105, thereby enabling wiring from the bracket 105. The lens can be restrained by abutting through the mounting hole; alternatively, the lens can be restrained by bonding the circuit board surface to the back of the bracket 105.

[0025] The field of view of an image acquisition device is determined by its field of view angle, viewing distance (or depth of field), and external influencing factors such as installation height and the presence of obstructions. The field of view can also be described as the range of scene space within which the image acquisition device can acquire effective images. The field of view of an image acquisition device refers to the angle formed by the lens of the device and the two edges of the target object that allow it to pass through the lens to its maximum extent. The field of view angle is typically between 30 and 120 degrees. A larger field of view angle results in a wider field of view. The resolution of an image acquisition device refers to the number of pixels it can capture; higher resolution results in a clearer image. Different image acquisition devices have different field of view angles and resolutions, allowing for a balance between a large field of view and high resolution in image acquisition. For example, the field of view of a first image acquisition device can be larger than that of a second image acquisition device, and the resolution of the second image acquisition device can be greater than that of the first. This allows the first image acquisition device to acquire a large area of ​​image information, including the target object, while the second image acquisition device can acquire a smaller, high-resolution image of the target object, enabling the acquisition of high-resolution features.

[0026] Based on the shooting requirements, the image acquisition components are designed and selected, determining the field of view and resolution of each component, as well as their relative positions, which in turn determines the location of the mounting holes. Since assembly errors are involved when the camera and bracket 105 are assembled together, these errors can be mitigated by performing overall calibration after assembling the camera onto the bracket 105 as a single unit. This calibration uniquely determines the relative positions of the fields of view of different image acquisition components. Then, after the first image acquisition component acquires image information with a large field of view, based on the target object's position in the image and the relative relationship between the fields of view of the second and first image acquisition components, an arm can be used to accurately move the target object into the field of view of the second image acquisition component, allowing the second component to capture the target object with a smaller field of view. This utility model's visual inspection module 100 is sold as a whole, avoiding errors caused by individually installing different image acquisition components, as well as the errors and inconvenience of calibrating each component after installation.

[0027] Depending on the required shooting accuracy, the number of image acquisition units can be two, three, or even more. The optical axes of the image acquisition units are parallel, ensuring that the angle of the target object acquired by different image acquisition units is consistent. Due to the different positions of the image acquisition units, their fields of view may or may not overlap. It is understood that in embodiments where the fields of view of different image acquisition units do not overlap, the relative positions of the fields of view of each image acquisition unit are known. The optimal shooting distance of different image acquisition units is consistent, thus eliminating the need to adjust the distance between the target object and the visual detection module 100 each time the target object is moved, thereby reducing movement deviation. The shooting process will be described later with reference to more specific embodiments.

[0028] This invention discloses a visual inspection module, a robot head device, and a robot. By setting up multiple image acquisition components with different fields of view and resolutions, it first acquires a wide range of image information using the large field-of-view component. Then, it determines the position of a target object from the image information. Next, it moves the target object into the field of view of the higher-resolution image acquisition component, continuing image acquisition and positioning. This process of acquiring higher-resolution image information one by one allows for rapid adjustment of the target object into the field of view of the high-resolution image acquisition component, achieving high-definition image information acquisition and improving the accuracy and efficiency of detection. Furthermore, by treating the visual inspection module as a whole and fixing the relative positions of the image acquisition components with a bracket, the relative relationship of the field of view positions of the image acquisition components is unique, resulting in high accuracy. Compared to installing individual image acquisition components onto the robot, this invention eliminates the need for user calibration.

[0029] In one embodiment, the plurality of image acquisition devices include a wide-angle camera lens module 101, a standard camera lens module 102, and a telephoto camera lens module 103. The field of view of the wide-angle camera lens module 101 is greater than that of the standard camera lens module 102 and the telephoto camera lens module 103. The resolution of the telephoto camera lens module 103 is greater than that of the wide-angle camera lens module 101 and the standard camera lens module 102.

[0030] The system uses three image acquisition units, which can be arranged in a triangle. For example, in practical use, the orientation might be... Figure 2As shown, the wide-angle camera lens module 101 and the standard camera lens module 102 are positioned at the same height at a higher position. The telephoto camera lens module 103 is positioned below the wide-angle camera lens module 101 and the standard camera lens module 102, and between them, the wide-angle camera lens module 101, the standard camera lens module 102, and the telephoto camera lens module 103 form an inverted triangle arrangement, thereby making the wide-angle camera lens module 101, the standard camera lens module 102, and the telephoto camera lens module 103 form an inverted triangle arrangement. The close proximity of the lens modules 103 reduces the angle deviation between the wide-angle camera lens module 101, the standard camera lens module 102, and the telephoto camera lens module 103 when shooting the target object, making the angle of the captured image more consistent. On the other hand, it ensures that the fields of view of the standard camera lens module 102 and the telephoto camera lens module 103 fall within the wide-angle camera lens module 101, and the field of view of the telephoto camera lens module 103 falls within the standard camera lens module 102, thus forming a nested field of view and reducing the distance the arm needs to move for each adjustment.

[0031] The wide-angle camera lens module 101 is used to acquire images with a large field of view and a large depth of field, enabling the capture of targets over a wide area. The standard camera lens module 102 is used to acquire standard images of the target object; its field of view is smaller than that of the wide-angle camera lens module 101, thus narrowing the range of the target object. The telephoto camera lens module 103 is a high-resolution camera with a small field of view, used to acquire high-resolution images of key areas or suspected defect areas of the target object. The optimal working distance of the wide-angle camera lens module 101, the standard camera lens module 102, and the telephoto camera lens module 103 is kept consistent, with the field of view decreasing, the depth of field decreasing, and the resolution increasing sequentially. In particular, the corresponding camera can be replaced according to different detection needs. In this embodiment, combined with the robot's agility, the images acquired by the wide-angle camera lens module 101 are used for target object identification and positioning, the images acquired by the standard camera lens module 102 are used for initial defect detection, and the images acquired by the telephoto camera lens module 103 are used for secondary judgment of suspected defects. This is a flexible, compact, and efficient appearance defect detection system.

[0032] To improve ambient brightness, in one embodiment, the visual detection module 100 further includes a supplementary light 104, which provides ambient light. Whether or not to provide lighting can be controlled according to the environment and shooting requirements.

[0033] The structure of the supplementary light 104 can be varied. For example, it can be multiple supplementary lights 104 arranged around the image acquisition components, or even four. By activating supplementary lights 104 at different positions, image information of the target object illuminated at different angles can be captured. Alternatively, the supplementary light 104 can be a ring light, surrounding the periphery of multiple image acquisition components. The supplementary light 104 provides uniform ambient light in the circumferential direction. The supplementary light 104 can be a single ring light or multiple ring lights, arranged axially spaced around the same axis. The inner diameters of the different ring lights are different, allowing different positions and numbers of ring lights to be activated as needed to provide a more suitable lighting environment.

[0034] In one embodiment, the support 105 includes a bottom wall and side walls, the side walls surrounding the bottom wall and extending to one side of the bottom wall, and the image acquisition element is disposed on the bottom wall. The bottom wall may be circular, and the side walls form an approximately cylindrical structure around the bottom wall.

[0035] The inner periphery of the sidewall includes an annular inclined surface extending toward the center of the area enclosed by the sidewall and simultaneously toward the bottom wall. The fill light 104 is disposed on the annular inclined surface, thereby making the fill light 104 closer to the target object and the light more focused, providing higher brightness with the same energy consumption and heat generation.

[0036] In one embodiment, the device further includes a face protector 106, which is made of a transparent material and covers the outer periphery of all image acquisition components. As in the aforementioned embodiment where the support 105 includes a bottom wall and a side wall, the face protector 106 covers the side wall opposite to the bottom wall.

[0037] The face protection component 106 can be high-transparency glass with an anti-reflective coating on its surface, used to protect the image acquisition component and the fill light 104, and also to prevent dust.

[0038] On the other hand, such as Figure 1-3 As shown, this embodiment of the present invention also provides a robot head device, including a vision detection module 100 as described above, and a depth camera 200, the depth camera 200 being used to detect and acquire depth information. A support component 300, the vision detection module 100, and the depth camera 200 are all connected to the support component 300.

[0039] The depth camera 200 can be any type of 3D camera, capable of acquiring precise distance and depth information of objects in the scene. For example, a binocular stereo vision camera can be used, capturing different perspectives of the scene through two cameras and estimating depth by calculating the parallax between the images. Using the depth camera 200, on the one hand, it can capture the three-dimensional shape and contour of objects for object recognition and classification, aiding in the identification of logical and structural anomalies. On the other hand, depth information helps adjust the distance between the object and the vision inspection module 100, ensuring that the object, or identified defects on the object, are within the optimal distance or depth-of-field coverage of the image acquisition unit. The depth camera 200 can be positioned on the top side of the vision inspection module 100, adjacent to it, to minimize positional and angular deviations.

[0040] The support assembly 300 is used to connect the robot head device to the robot body. The support assembly 300 may simply provide connection and support, or it may have angular and directional adjustment functions. In one embodiment, the support assembly 300 includes a fixing member 301 and a neck connection mechanism 302. The vision inspection module 100 and the depth camera 200 are both connected to the fixing member 301. The neck connection mechanism 302 includes a first adjustment assembly and a second adjustment assembly, the first adjustment assembly being connected to the second adjustment assembly, and the second adjustment assembly being connected to the fixing member 301. The first adjustment assembly drives the second adjustment assembly and the fixing member 301 to rotate around a rotation axis, and the second adjustment assembly drives the fixing member 301 to swing relative to a swing axis.

[0041] The robot head device includes an embodiment of the vision detection module 100 as described above, and the advantages of the aforementioned embodiment of the vision detection module 100 will not be repeated here.

[0042] The fixing component 301 can be simply a support frame, or it can be a hollow box structure. The first adjustment assembly is located on the lowest side and may include a base and a rotary motor. The base connects to the robot's main body, and the rotary motor's body is connected to the base. The second adjustment assembly includes a frame, a swing motor, and a swing arm. The frame is connected to the output end of the rotary motor, the swing motor's body is connected to the frame, the swing motor's output end is connected to the first end of the swing arm, and the second end of the swing arm is connected to the fixing component 301. By controlling the rotary motor and the swing motor, the fixing component 301 and the connected vision detection module 100 and depth camera 200 can achieve two degrees of freedom: pitch and head tilt. The head device's posture can be adjusted in real time, either actively or through external signal interaction. Figure 3 As shown, the pitch adjustment is from eye level to a 30° head-down angle, as... Figure 3In the center, when looking straight ahead, the optical axis of the image acquisition device is horizontal, meaning the field of view of the image acquisition device changes from area A to area B. The head tilt adjustment range is -45° to 45°, meaning it can be adjusted to any angle position between 45° to the left and 45° to the right.

[0043] In one embodiment, the device further includes an interactive element 400, connected to the support component 300, for receiving sound signals and generating feedback signals. The device also includes a controller 600, connected to the support component 300.

[0044] The controller 600 is electrically connected to the interactive element 400, the image acquisition unit and supplementary light 104 of the vision detection module 100, and the depth camera 200. On one hand, it integrates and calculates the image information acquired by the image acquisition unit and the depth camera 200 to determine the position of the target object, and then sends a feedback signal to the robot body to control the movement of the robot's arm. It also controls the image acquisition unit to capture images and the on / off state and brightness adjustment of the supplementary light 104. Furthermore, it controls the neck connection mechanism 302 to adjust the pitch and swing angles. On the other hand, it controls the robot according to a pre-set program or instructions input by the user through the interactive element 400. In one embodiment, the interactive element 400 is a voice device that can acquire the user's voice information. The controller 600 receives the feedback signal converted from the voice information acquired by the interactive element 400, and then controls the image acquisition unit to capture images and the on / off state and brightness adjustment of the supplementary light 104 based on external active input. It also controls the neck connection mechanism 302 to adjust the pitch and swing angles.

[0045] The number of interactive components 400 can be one or more, and multiple interactive components 400 can be set on different sides of the fixing component 301 to achieve the acquisition of interactive signals from different sides. Interactive components 400 can have multiple functions; for example, they can integrate modules such as wireless WiFi, audio, speakers, and amplifiers. In addition to the aforementioned voice control, they can also announce the image acquisition process and perform defect identification. Alternatively, in some embodiments, the interactive component 400 can be simply an amplifier, and the wireless WiFi, audio, and speakers can be set within the controller 600, thereby reducing the space occupied by the interactive component 400.

[0046] In one embodiment, the device further includes a heat dissipation component 500, which is directly or indirectly connected to the controller 600 and is used to dissipate heat from the controller 600.

[0047] In embodiments where the fixing member 301 is merely a support, the controller 600 can be positioned above the support, and the heat dissipation assembly 500 can include multiple parallel heat dissipation fins connected to the housing of the controller 600 for heat dissipation. Alternatively, in embodiments where the fixing member 301 is a hollow structure, the controller 600 can be housed within the fixing member 301, and the heat dissipation assembly 500 can include multiple parallel heat dissipation fins connected to the side wall of the fixing member 301. A fan can also be installed within the fixing member 301 to promote airflow within the fixing member 301, thereby dissipating heat from the controller 600.

[0048] On another front, this utility model discloses a robot, including a vision detection module 100 as described above, and a main body, with a support 105 connected to the main body, the main body including at least an arm for picking up and moving a target object. Alternatively, it includes a robot head device as described above, and a main body, with a support component 300 connected to the main body, the main body including at least an arm for picking up and moving a target object.

[0049] The robot includes any of the above-described embodiments of the vision inspection module 100 or robot head device, and the advantages of including any of the aforementioned vision inspection module 100 or robot head device embodiments will not be elaborated here.

[0050] The arm is used to pick up and move the target object. The following is a detailed description of an image acquisition process, in conjunction with the aforementioned embodiments: S1. The robotic arm grasps the target object and adjusts the posture of the robot head device through the support component 300 until the target object's shooting point 1 appears in the field of view of the wide-angle camera lens module 101 and within the depth of field coverage, a large field of view and high depth of field image of point 1 is obtained. S2. By performing real-time calculations on the images acquired by the wide-angle camera lens module 101, the robot arm adjusts the target object to be within the field of view of the standard camera lens module 102 and within the depth of field coverage. S3. By performing real-time calculations on the images acquired by the standard camera lens module 102, the robot arm, holding the target object, is adjusted back into the field of view of the telephoto camera lens module 103, and a high-resolution image of point 1 is obtained within the depth of field coverage.

[0051] S4. Adjust the robot head device posture to the next shooting point of the target object through the support component 300, and perform image acquisition by the wide-angle camera lens module 101. Repeat S2 to S3 until all points of the target object have been imaged.

[0052] The above description is merely a specific embodiment of this utility model, but the protection scope of this utility model is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this utility model should be included within the protection scope of this utility model. Therefore, the protection scope of this utility model should be determined by the protection scope of the claims.

Claims

1. A visual inspection module, characterized in that, For use in robots, the vision detection module (100) includes: Multiple image acquisition devices, each with a different field of view and resolution, are used to acquire image information of the target object; The image acquisition element is mounted on the bracket (105) to fix the relative positions of the different image acquisition elements.

2. The visual inspection module according to claim 1, characterized in that, The optical axes of the different image acquisition devices are parallel, and the fields of view of the different image acquisition devices have overlapping areas, and the optimal shooting distance of the different image acquisition devices is the same.

3. The visual inspection module according to claim 1, characterized in that, The plurality of image acquisition devices include a wide-angle camera lens module (101), a standard camera lens module (102), and a telephoto camera lens module (103). The field of view of the wide-angle camera lens module (101) is greater than that of the standard camera lens module (102) and the telephoto camera lens module (103); The resolution of the telephoto camera lens module (103) is greater than that of the wide-angle camera lens module (101) and the standard camera lens module (102).

4. The visual inspection module according to claim 1, characterized in that, The visual detection module (100) also includes a supplementary light (104) for providing ambient light.

5. The visual inspection module according to claim 4, characterized in that, The fill light (104) is a ring light, and the fill light (104) surrounds the periphery of the plurality of image acquisition components.

6. The visual inspection module according to claim 5, characterized in that, The bracket (105) includes a bottom wall and a side wall, the side wall surrounds the bottom wall and extends to one side of the bottom wall, and the image acquisition element is disposed on the bottom wall; The inner periphery of the sidewall includes an annular inclined surface extending toward the center of the area enclosed by the sidewall and simultaneously toward the bottom wall, and the supplementary light (104) is disposed on the annular inclined surface.

7. The visual inspection module according to claim 1, characterized in that, The bracket (105) includes a bottom wall and a side wall, the side wall surrounds the bottom wall and extends to one side of the bottom wall, and the image acquisition element is disposed on the bottom wall; It also includes a face protector (106), which is made of a transparent material and covers the side of the sidewall opposite to the bottom wall.

8. A robot head device, characterized in that, Includes the visual detection module (100) as described in any one of claims 1-7, and A depth camera (200) is used to detect and acquire depth information; The support component (300) is connected to both the visual inspection module (100) and the depth camera (200).

9. The robot head device according to claim 8, characterized in that, The device further includes: An interactive element (400) is connected to the support component (300) and is used to receive sound signals and generate feedback signals.

10. The robot head device according to claim 8, characterized in that, The support assembly (300) includes a fastener (301) and a neck connection mechanism (302). Both the visual inspection module (100) and the depth camera (200) are connected to the fixture (301); The neck connection mechanism (302) includes a first adjustment component and a second adjustment component, the first adjustment component being connected to the second adjustment component, and the second adjustment component being connected to the fastener (301); The first adjustment component is used to drive the second adjustment component and the fixing member (301) to rotate around the rotation axis, and the second adjustment component is used to drive the fixing member (301) to swing relative to the swing axis.

11. The robot head device according to claim 8, characterized in that, The device also includes a controller (600) connected to the support assembly (300); The device further includes a heat dissipation component (500), which is directly or indirectly connected to the controller (600) and is used to dissipate heat from the controller (600).

12. The robot head device according to claim 11, characterized in that, The heat dissipation assembly (500) includes a fan and / or multiple heat dissipation fins arranged in parallel.

13. A robot, characterized in that, The device includes a visual detection module (100) as described in any one of claims 1-7 above, and a body, wherein the support (105) is connected to the body, and the body includes at least an arm for picking up and moving a target object; Alternatively, it may include a robot head assembly as described in any one of claims 8-12 above, and a body to which the support assembly (300) is connected, the body including at least an arm for picking up and moving a target object.

Images