Visual terminal for AI face recognition, locking and tracking

Abstract

The utility model relates to a visual terminal for AI face recognition, locking and tracking. The visual terminal comprises a base, a lifting mechanism, a bearing platform and a camera. The lifting mechanism is installed on the base and used for driving the bearing platform to ascend and descend. The bearing platform comprises a connecting frame, a movable plate and a pitching mechanism; the connecting frame is connected to the lifting mechanism, the movable plate is rotationally connected to the connecting frame, and the pitching mechanism is used for driving the movable plate to rotate relative to the connecting frame; and the camera is mounted on the movable plate. The lifting mechanism and the pitching mechanism are started according to the height of the target in cooperation with a software algorithm, so that the visual angle of the camera is at a better height and angle, and the problems that the position of a traditional camera is fixed and the target can avoid camera locking and tracking when lowering the head are solved. Therefore, the problem that it is difficult to identify and track the target in an environment with various height changes is solved.

Description

Technical Field

[0001] This application relates to the field of video tracking technology, and in particular to a visual terminal for AI face recognition and tracking. Background Technology

[0002] In recent years, driven by the rapid advancements in artificial intelligence and deep learning technologies, facial recognition technology has made significant progress and is gradually maturing. Numerous internet companies, such as Baidu, Tencent, and Alibaba, have actively invested in the research and application of facial recognition, indicating that this technology is about to enter a new stage of large-scale industrialization. This technological innovation not only heralds a profound transformation of "AI + vertical industries" but also brings unprecedented opportunities for intelligent upgrading to all sectors of society.

[0003] Currently, camera systems based on AI facial recognition tracking technology can quickly identify and track targets within their field of view. However, in practical applications, traditional cameras are usually installed at fixed heights, a layout that proves inadequate when faced with complex and ever-changing environmental requirements. Especially in scenarios where adults and young children coexist, or where the target frequently changes their walking posture (e.g., walking with their head down), the significant differences in required recognition and tracking heights often make it difficult for existing camera systems to effectively handle these situations, leading to challenges in identifying and tracking targets in environments with varying heights. Utility Model Content

[0004] To address the challenge of identifying and tracking targets in highly variable environments, this application provides an AI-powered facial recognition-based visual terminal for tracking and locking targets.

[0005] The visual terminal for AI face recognition locking and tracking provided in this application adopts the following technical solution:

[0006] A visual terminal for AI face recognition and tracking includes a base, a lifting mechanism, a support platform, and a camera. The lifting mechanism is mounted on the base and is used to lift the support platform. The support platform includes a connecting frame, a movable plate, and a tilting mechanism. The connecting frame is connected to the lifting mechanism, the movable plate is rotatably connected to the connecting frame, and the tilting mechanism is used to drive the movable plate to rotate relative to the connecting frame. The camera is mounted on the movable plate.

[0007] By adopting the above technical solution, a lifting mechanism is used to drive the platform carrying the camera to rise and fall, thereby changing the camera's vertical field of view. Combined with a tilt mechanism, the camera's observation angle is changed. Thus, when AI recognizes and tracks faces, the lifting and tilt mechanisms are activated based on the target's height and software algorithm, so that the camera's field of view is at a better height and angle. This improves the problem of traditional cameras being fixed in position and the target being able to avoid camera lock-on tracking when looking down, thereby improving the difficulty of recognizing and tracking targets in environments with varying heights.

[0008] Optionally, the pitch mechanism includes a sector gear, a drive gear, and a servo motor; the sector gear is mounted on the movable plate, the drive gear is coaxially fixed on the drive shaft of the servo motor, and the drive gear meshes with the sector gear; the servo motor is fixed on the connecting frame.

[0009] By adopting the above technical solution, on the one hand, the interaction between the sector gear and the drive gear enables the movable plate to rotate relative to the lifting mechanism, thereby achieving the effect of adjusting the camera's tilt angle. On the other hand, the servo motor can precisely adjust the rotation angle and torque control, with a fast response speed. During the process of recognizing and tracking faces, it can adjust the camera's tilt angle in a timely manner to adapt to changes in the target person's posture, such as looking up or down.

[0010] Optionally, both the sector gear and the drive gear are located directly below the camera.

[0011] By adopting the above technical solution, the design of the sector gear and drive gear located directly below the camera can ensure that the meshing surface of the gear remains perpendicular to the rotation axis of the movable plate during rotation, which helps to reduce errors caused by gear tilting and thus maintain high transmission accuracy.

[0012] Optionally, a protective cylinder is also included, which is connected to the base, and the lifting mechanism, the supporting platform and the camera are all located inside the protective cylinder.

[0013] By adopting the above technical solutions, the adhesion of external pollutants to the lifting mechanism, pitching mechanism, and camera is reduced, thereby improving their service life.

[0014] Optionally, the protective cylinder has a window on its side wall, and the lens of the camera faces the window; the protective cylinder has glass installed on its side wall, and the glass covers the window.

[0015] By adopting the above technical solutions, the external perception of the camera is reduced, while the protective casing's protective capabilities are further improved.

[0016] Optionally, the connecting frame is equipped with an elastic damping block, which abuts against the surface of the glass.

[0017] By adopting the above technical solution and introducing the elastic shock absorber, on the one hand, the lifting mechanism absorbs the vibration of the connecting frame during the lifting of the carrying platform, thus improving the problem of image shaking during the lifting process of the camera; on the other hand, the elastic shock absorber wipes the surface of the glass, reducing the labor intensity of manual glass wiping.

[0018] Optionally, a first magnetic block is installed on the connecting frame; a wiping block is abutted against the outer wall of the glass outside the protective cylinder, and a second magnetic block is installed on the wiping block, the second magnetic block and the first magnetic block attract each other.

[0019] By adopting the above technical solution, and utilizing a wiping block, the mutual attraction between the first and second magnetic blocks during the lifting and lowering of the connecting frame achieves the effect of automatically wiping the glass surface, reducing the risk of external contaminants obstructing the camera and causing face tracking failure. Furthermore, the addition of elastic shock-absorbing blocks enables automatic wiping of both the inner and outer surfaces of the glass.

[0020] Optionally, the base is rotatably connected to a turntable away from the center of the bottom surface of the lifting mechanism.

[0021] By adopting the above technical solution and introducing the turntable, the base can be quickly rotated during the initial deployment or mid-term adjustment to rapidly adjust the 360° observation position of the camera.

[0022] In summary, this application includes at least one of the following beneficial technical effects:

[0023] By using a lifting mechanism, the platform carrying the camera is raised and lowered to change the camera's vertical field of view. Combined with a tilt mechanism, the camera's viewing angle is changed. Thus, when AI recognizes and tracks faces, the lifting and tilt mechanisms are activated based on the target's height and software algorithm to ensure that the camera's field of view is at an optimal height and angle. This improves upon the problem of traditional cameras being fixed in position and targets being able to avoid camera lock-on tracking when they look down, thereby improving the difficulty of recognizing and tracking targets in environments with varying heights.

[0024] By using a wiping block and an elastic shock-absorbing block, the first and second magnetic blocks attract each other during the lifting and lowering of the connecting frame, the inner and outer surfaces of the glass are automatically wiped, reducing the risk of external pollutants obscuring the camera and causing face tracking failure.

[0025] The introduction of the turntable allows for rapid adjustment of the camera's 360° observation position during initial deployment or mid-term adjustments by quickly rotating the base. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall structure of an embodiment of this application.

[0027] Figure 2 This is a schematic diagram used to show the state of the glass separating from the protective casing.

[0028] Figure 3 yes Figure 2 Enlarged view of section A.

[0029] Figure 4 This is a schematic diagram showing the installation position of the second magnetic block.

[0030] Explanation of reference numerals in the attached drawings: 1. Base; 2. Lifting mechanism; 3. Supporting platform; 31. Connecting frame; 32. Movable plate; 33. Tilting mechanism; 331. Sector gear; 332. Drive gear; 333. Servo motor; 4. Camera; 5. Protective cylinder; 51. Window; 52. Glass; 6. Turntable; 7. Elastic shock absorber; 8. First magnetic block; 9. Wiping block; 10. Second magnetic block. Detailed Implementation

[0031] The following is in conjunction with the appendix Figure 1-4 This application will be described in further detail.

[0032] This application discloses a visual terminal for AI face recognition and tracking.

[0033] Reference Figure 1 and Figure 2 A visual terminal for AI face recognition and tracking includes a base 1, a lifting mechanism 2, a support platform 3, a camera 4, and a protective cylinder 5. The base 1 is frustum-shaped, with a turntable 6 rotatably connected to the center of its bottom surface. The lifting mechanism 2 is mounted on the top surface of the base 1 and is used to lift the support platform 3. The camera 4 is mounted on the support platform 3. The protective cylinder 5 is connected to the base 1, and the lifting mechanism 2, support platform 3, and camera 4 are all located inside the protective cylinder 5. The turntable 6 allows the base 1 to rotate omnidirectionally, thereby enabling omnidirectional adjustment of the camera 4's observation range. The lifting mechanism 2, in conjunction with the support platform 3, allows for vertical adjustment of the camera 4's observation range.

[0034] Reference Figure 1 and Figure 2Specifically, the rotational connection between the base 1 and the turntable 6 can be a structure using a rotating shaft and bearings, or a configuration combining a motor and bearings. This means that this connection method can achieve rotation under manual operation as well as automated rotation through system control. The lifting mechanism 2 is a commonly used component in the field of mechanical manufacturing. This application uses a lead screw and slider mechanism. The specific structure and working principle of the lead screw and slider mechanism will not be described in detail in this application.

[0035] Reference Figure 3 The carrying platform 3 includes a connecting frame 31, a movable plate 32, and a pitching mechanism 33. The connecting frame 31 is bolted to the slider of the lifting mechanism 2. The movable plate 32 is rotatably connected to the connecting frame 31. The pitching mechanism 33 is used to drive the movable plate 32 to rotate relative to the connecting frame 31.

[0036] Reference Figure 3 The pitch mechanism 33 includes a sector gear 331, a drive gear 332, and a servo motor 333. The sector gear 331 is fixed to the surface of the movable plate 32 facing away from the camera 4. The drive gear 332 is coaxially fixed to the drive shaft of the servo motor 333. The drive gear 332 meshes with the sector gear 331. Both the sector gear 331 and the drive gear 332 are located directly below the camera 4. The servo motor 333 is bolted to the connecting bracket 31.

[0037] Reference Figure 3 and Figure 4 The protective cylinder 5 has a window 51 on its side wall, and the lens of the camera 4 faces the window 51. A glass 52 is bolted to the side wall of the protective cylinder 5, covering the window 51. An elastic shock-absorbing block 7 is detachably connected to the connecting frame 31; the detachable connection method can be plug-in, bolted, or snap-fit. A first magnetic block 8 is adhered to the connecting frame 31, and the elastic shock-absorbing block 7 covers the first magnetic block 8. The elastic shock-absorbing block 7 abuts against the inner surface of the glass 52 located on the protective cylinder 5. A wiping block 9 abuts against the outer surface of the glass 52 located outside the protective cylinder 5. A second magnetic block 10 is adhered to the wiping block 9, and the second magnetic block 10 attracts the first magnetic block 8.

[0038] The implementation principle of an AI face recognition and tracking visual terminal according to this application embodiment is as follows: When the AI ​​recognizes and tracks a face, based on the target's height and the software algorithm, the lifting mechanism 2 and the tilt mechanism 33 are activated. The lifting mechanism 2 drives the support platform 3 carrying the camera 4 to rise and fall, and the tilt mechanism 33 changes the viewing angle of the camera 4, so that the camera 4's viewing angle is at an optimal height and angle. During the lifting and lowering process of the support platform 3, the elastic shock-absorbing block 7 improves the sliding problem of the camera 4 during movement, and together with the wiping block 9, it achieves the effect of automatically wiping the inner and outer surfaces of the glass 52, reducing the risk of external pollutants obscuring the camera 4 and causing face tracking failure. This improves the problem of the traditional fixed position of the camera 4, which can avoid the camera 4 when the target lowers its head for tracking, thereby improving the difficulty of recognizing and tracking targets in environments with varying heights.

[0039] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A visual terminal of AI face recognition lock tracking, characterized in that: It includes base (1), lifting mechanism (2), bearing platform (3) and camera (4), lifting mechanism (2) is installed on the base (1), lifting mechanism (2) is used to drive bearing platform (3) to lift, bearing platform (3) includes connecting frame (31), movable plate (32) and pitching mechanism (33), connecting frame (31) is connected on lifting mechanism (2), movable plate (32) is rotatably connected on connecting frame (31), pitching mechanism (33) is used to drive movable plate (32) to rotate relative to connecting frame (31), camera (4) is installed on movable plate (32), base (1) is rotatably connected with turntable (6) in the center of the bottom surface away from lifting mechanism (2). 2.The visual terminal of claim 1, wherein: The pitching mechanism (33) includes a sector gear (331), a drive gear (332) and a servo motor (333), the sector gear (331) is installed on the movable plate (32), the drive gear (332) is coaxially fixed on the drive shaft of the servo motor (333), and the drive gear (332) is engaged with the sector gear (331), the servo motor (333) is fixed on the connecting frame (31). 3.The visual terminal of claim 2, wherein: The sector gear (331) and the drive gear (332) are located directly below the camera (4).

4. The visual terminal of claim 1, wherein: It also includes a protective cylinder (5), which is connected to the base (1), the lifting mechanism (2), the bearing platform (3) and the camera (4) are located in the protective cylinder (5). 5.The visual terminal of claim 4, wherein: The sidewall of the protective cylinder (5) is provided with a window (51), and the lens of the camera (4) faces the window (51), the sidewall of the protective cylinder (5) is provided with a glass (52), and the glass (52) covers the window (51). 6.The visual terminal of claim 5, wherein: The connecting frame (31) is provided with an elastic damping block (7), and the elastic damping block (7) abuts against the surface of the glass (52). 7.The visual terminal of claim 5 or 6, wherein: The connecting frame (31) is provided with a first magnetic block (8), the outer sidewall of the protective cylinder (5) is provided with a wiping block (9), the wiping block (9) is provided with a second magnetic block (10), and the second magnetic block (10) and the first magnetic block (8) are attracted to each other.

Images