Automatic tracking and monitoring system for spreaders based on AI recognition
By using an AI-recognition-based automatic tracking and monitoring system in the monitoring system, combined with the stepper motor and transmission mechanism on the gimbal, a stable and accurate monitoring picture is achieved in the vibration environment, solving the problem of failure of the monitoring system in the existing technology in the vibration environment, and improving the efficiency and safety of the operating process.
Patent Information
- Application Number
- CN202421343049.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-13
- Publication Date
- 2025-04-04
- Estimated Expiration
- 2034-06-13
AI Technical Summary
The existing monitoring system cannot provide a stable and accurate monitoring screen that follows the position of the observed object in a vibrating environment, and the gimbal needs to rely on PLC information for control, which is limited by whether the main machine provides the corresponding data signal.
The automatic tracking and monitoring system of spreader based on AI recognition is adopted to control the camera system for tracking through AI recognition. The stepper motor, gear transmission pair, worm and worm gear transmission on the gimbal is used, and the photoelectric switch and control circuit board are combined to achieve accurate positioning and tracking of the camera device.
Provide a stable and precise camera system that follows the position of the observed object in a vibrating environment, improving the efficiency and safety of the operation process and avoiding excessive dependence on PLC signals.
Smart Images

Figure CN222714957U_ABST
Abstract
Description
Technical Field
[0001] The utility model relates to the technical field of video monitoring, and in particular to an automatic tracking and monitoring system for a sling based on AI recognition. Background Art
[0002] As an important port industry, ship unloaders and other operating areas need to be closely monitored to effectively ensure safe and orderly production, detect abnormal conditions in a timely manner, and stop abnormal processes in the bud.
[0003] In the existing monitoring system, the technical solutions mostly adopt the following two methods:
[0004] 1. Dome camera: This type of equipment is usually driven by synchronous belts or gears, which are prone to losing steps in a vibrating environment. Its power source is usually a stepper motor, which has a torque of almost zero after power failure. This means that once the device loses power in a vibrating environment, the device position will inevitably change. At the same time, the stepper motor does not have the function of detecting and compensating for lost steps. Therefore, dome cameras cannot provide stable and accurate monitoring images that follow the position of the observed object in a vibrating environment.
[0005] 2. A PTZ machine is used, but it needs to obtain the information of the ship unloader hoisting equipment, communicate with the PLC, and then control it. This is limited by whether the large machine can provide the corresponding data signal, which affects the implementation of the project. Utility Model Content
[0006] The purpose of the utility model is to address the defects and shortcomings of the prior art and provide an automatic tracking and monitoring system for slings based on AI recognition, which controls the camera system for tracking through AI recognition and can provide a stable and accurate camera system that follows the position changes of the observed object in a vibrating environment, making the operation process more efficient and safer.
[0007] To achieve the above-mentioned purpose, the utility model adopts the following technical solutions: it comprises a pan-tilt, a camera device and a software platform; the pan-tilt is connected with the camera device; the software platform and the pan-tilt are connected with a signal transmission; the pan-tilt comprises:
[0008] A shell, wherein two ends of the shell are respectively connected to a front cover and a rear cover;
[0009] A stepper motor is fixed inside the housing; and an output end of the stepper motor is connected to a gear transmission pair;
[0010] A worm gear fixed to a driven gear in a gear transmission pair;
[0011] The main shaft is rotatably mounted inside the housing by means of a bearing, one end of which is rotatably connected to the front end cover by means of a bearing, and is arranged to penetrate inside the main shaft; one end of the main shaft located inside the housing is connected to a photoelectric sensor sheet matched with a photoelectric switch; the photoelectric switch is fixed inside the housing; the camera device is connected to the outer end of the main shaft by means of a coupling;
[0012] A worm wheel, which is sleeved on the main shaft and is connected to the worm gear;
[0013] A control circuit board is arranged inside the shell and fixed on the rear end cover; the control circuit board is electrically connected to the stepping motor and the photoelectric switch.
[0014] Preferably, the camera device comprises a movement, the movement is arranged in an outer tube, the outer wall of the outer tube is connected to the coupling; one end of the outer tube is connected to a back cover, and the other end of the outer tube is provided with tempered glass.
[0015] Preferably, the outer cover of the tempered glass is provided with a sunshade, and the sunshade is connected to the end of the outer tube.
[0016] Compared with the prior art, the beneficial effects of the utility model are as follows: the utility model provides an automatic tracking and monitoring system for slings based on AI recognition, which controls the camera system for tracking through AI recognition, and can provide a stable and accurate camera system that follows the position changes of the observed object in a vibrating environment, making the operation process more efficient and safer. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a control block diagram of the utility model.
[0018] Figure 2 It is a schematic diagram of the connection between the pan / tilt head and the camera device in the utility model.
[0019] Figure 3 It is a structural schematic diagram of the pan / tilt platform in the utility model.
[0020] Figure 4 It is a structural schematic diagram of the camera device in the utility model.
[0021] Figure 5 It is an end view of the camera device in the utility model.
[0022] Figure 6 It is a power supply circuit diagram in the utility model.
[0023] Figure 7 It is a CPU control circuit diagram in the utility model.
[0024] Figure 8 It is a communication interface circuit diagram in the utility model.
[0025] Fig. 9 This is a reference detection circuit diagram in the utility model.
[0026] Fig.10 This is a stepper motor control circuit diagram in the utility model.
[0027] Description of reference numerals:
[0028] Pan / tilt head 1, housing 1-1, stepper motor 1-2, worm 1-3, main shaft 1-4, worm gear 1-5, front cover 1-6, rear cover 1-7, control circuit board 1-8, photoelectric switch 1-9, camera device 2, rear cover 2-1, outer tube 2-2, movement 2-3, light shield 2-4, tempered glass 2-5, coupling 3, software platform 4. DETAILED DESCRIPTION
[0029] The technical solution of the utility model will be clearly and completely described below in conjunction with the accompanying drawings. The preferred embodiments in the description are only used as examples, and all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the utility model.
[0030] like Figure 1-Figure 5 As shown, this specific embodiment adopts the following technical scheme: it includes a pan-tilt 1, a camera device 2 and a software platform 4; both are made of stainless steel, and the manufacturing and material costs are relatively low; the pan-tilt 1 is connected to the camera device 2; the software platform 4 is connected to the pan-tilt 1 for signal transmission; the pan-tilt 1 includes:
[0031] A housing 1-1, wherein two ends of the housing 1-1 are respectively connected to a front cover 1-6 and a rear cover 1-7;
[0032] A stepper motor 1-2, wherein the stepper motor 1-2 is fixed inside the housing 1-1; and an output end of the stepper motor 1-2 is connected to a gear transmission pair (not shown in the figure);
[0033] A worm 1-3, wherein the worm 1-3 is fixed on a driven gear in a gear transmission pair;
[0034] The spindle 1-4 is rotatably mounted inside the housing 1-1 by means of a bearing, one end of which is rotatably connected to the front end cover 1-6 by means of a bearing, and the spindle 1-4 is arranged to penetrate inside; one end of the spindle 1-4 located inside the housing 1-1 is connected to a photoelectric sensor sheet matched with a photoelectric switch 1-9; the photoelectric switch 1-9 is fixed inside the housing 1-1; the camera device 2 is connected to the outer end of the spindle 1-4 by means of a coupling 3;
[0035] A worm wheel 1-5, wherein the worm wheel 1-5 is sleeved on the main shaft 1-4 and is drivingly connected to the worm 1-3;
[0036] A control circuit board 1-8, wherein the control circuit board 1-8 is arranged inside the housing 1-1 and fixed on the rear end cover 1-7; the control circuit board 1-8 is electrically connected to the stepping motor 1-2 and the photoelectric switch 1-9;
[0037] See also Figure 6-Figure 10 As shown, the control circuit board 1-8 is composed of a power supply circuit, a CPU control circuit, a communication interface circuit, a reference detection circuit, and a stepper motor control circuit. The input voltage of the power supply circuit is AC24V, which provides DC5V and DC3.3V voltages for the entire system. A self-recovery insurance is added to the input end for overcurrent protection; the CPU control circuit converts the instructions received from AI recognition into control instructions for the tracking camera system. The CPU control circuit uses the STC15W4K32S4 single-chip microcomputer as a processor. According to the position relationship of the sling in the picture in the AI recognition instruction, the rotation angle of the camera device 2 is calculated, and the angle is changed by controlling the stepper motor 1-2, and the distance between the tracked object and the camera. According to the focal length of the single-magnification image, the multiple that should be changed is calculated, and then the coordinate value of the zoom motor corresponding to the zoom factor is calculated; by collecting the current coordinates of the camera device 2, the difference between the predetermined coordinates and the current coordinates is calculated, and the speed to be selected is calculated. The degree level is sent to the camera device 2 together for zoom processing, so as to achieve synchronization and continuity between the movement of the tracked object and the zoom of the camera device 2; the communication interface circuit converts the TTL signal and the RS485 signal, and the interface is provided with a protection circuit composed of a self-recovery insurance and a TVS diode SMBJ5.0A, which can withstand the misconnection voltage of AC48V and has overvoltage and overcurrent protection functions; the reference detection circuit collects the reference signal to determine whether the reference point has been reached, and is used for 0-point reference calibration. After the tracking camera system is turned on, it will perform self-test, and finally calibrate by detecting the reference point; the stepper motor control circuit is used to control the stepper motor 1-2. When the position coordinate changes, the pan-tilt angle needs to change, and the camera device 2 reaches the corresponding angle by controlling the rotation of the stepper motor 1-2;
[0038] The camera device 2 includes a movement 2-3, which is arranged in an outer tube 2-2, and the outer wall of the outer tube 2-2 is connected to the coupling 3; one end of the outer tube 2-2 is connected to the back cover 2-1, and the other end of the outer tube 2-2 is provided with a tempered glass 2-5, and the outer cover of the tempered glass 2-5 is provided with a sunshade 2-4, and the sunshade 2-4 is connected to the end of the outer tube 2-2.
[0039] When using the utility model, the system is connected to an external power supply and reset through the photoelectric switch 1-9 inside the pan / tilt head 1. At the same time, the photoelectric switch 1-9 is used to promptly correct possible deviations that may occur during operation to ensure accurate positioning each time the control is performed and reduce errors. The software platform 4 acquires images in real time and performs AI recognition. After recognition, the position of the sling in the picture is framed and given to the movement 2-3. After receiving the command, the movement 2-3 sends the frame selection center coordinate information to the control circuit board 1-8. The control circuit board 1-8 calculates the angle that the camera device 2 needs to rotate according to the frame selection center coordinate, and then calculates the number of pulses of the stepper motor 1-2 to control the stepper motor 1-2 to rotate, and uses the gear transmission pair to drive the worm 1-3, thereby driving the worm wheel 1-5 to rotate, and the worm wheel 1-5 drives the main shaft 1-4 to rotate. The main shaft 1-4 uses the coupling 3 to drive the camera device 2 to rotate to the corresponding angle, so as to achieve the effect of real-time tracking.
[0040] The zoom processing of the camera device 2 is as follows: through the AI recognition software, it is set in advance that the proportion of the sling in the picture is, for example, one-half, and the number of pixels of the sling in the picture is calculated. If it is less than one-half, it is enlarged to achieve the effect of consistent proportion of the props; otherwise, it is reduced.
[0041] Compared with the prior art, the beneficial effects of the utility model are:
[0042] 1. The tracking camera system can track the object being tracked through AI recognition, while avoiding over-reliance on PLC information; it solves the problem that the existing PTZ machine can only track by obtaining PLC signals.
[0043] 2. High mechanical strength: The whole machine is made of stainless steel, each part is tightly connected, and the structure is good, which improves the overall strength of the gimbal while taking into account the sealing performance, and can cope with a variety of complex external environments.
[0044] 3. Stability: The pan / tilt adopts worm gear transmission, which improves the overall stability of the equipment and the stability of the system under vibration conditions, eliminates the influence of step loss on the positioning accuracy of the pan / tilt, and adds a photoelectric switch to the pan / tilt to make the information of the initial position of the pan / tilt more accurate. The difficulty of calculating the preset position is greatly reduced, thus improving the positioning accuracy of the pan / tilt.
[0045] 4. Low cost: The current transmission form of the existing pan / tilt is gear transmission. For processing, gear transmission requires high-precision mechanical equipment for processing, and the parts after processing need to be tempered or other heat treatment processes, so the manufacturing cost of the gear is high, and the assembly precision required for the gear matching is high during assembly, so it requires highly qualified personnel for installation and debugging, so the cost is high. This system uses a stepper motor to drive the gear, worm gear, and worm transmission, and the manufacturing and material costs are relatively low.
[0046] For those skilled in the art, they can modify the technical solutions recorded in the aforementioned embodiments and make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present utility model should be included in the protection scope of the present utility model.
Claims
1. The automatic tracking and monitoring system for spreaders based on AI recognition is characterized by: It comprises a pan-tilt platform (1), a camera device (2) and a software platform (4); both are made of stainless steel, and the manufacturing and material costs are relatively low; the pan-tilt platform (1) is connected to the camera device (2); the software platform (4) and the pan-tilt platform (1) are connected for signal transmission; the pan-tilt platform (1) comprises: A shell (1-1), wherein two ends of the shell (1-1) are respectively connected to a front end cover (1-6) and a rear end cover (1-7); A stepper motor (1-2), wherein the stepper motor (1-2) is fixed inside the housing (1-1); and an output end of the stepper motor (1-2) is connected to a gear transmission pair; A worm (1-3), wherein the worm (1-3) is fixed on a driven gear in a gear transmission pair; A main shaft (1-4), the main shaft (1-4) being rotatably mounted inside the housing (1-1) by means of a bearing, one end of the main shaft (1-4) being rotatably connected to the front end cover (1-6) by means of a bearing, and being arranged to penetrate inside the main shaft (1-4); one end of the main shaft (1-4) located inside the housing (1-1) being connected to a photoelectric sensing sheet matched with a photoelectric switch (1-9); the photoelectric switch (1-9) being fixed inside the housing (1-1); and the camera device (2) being connected to the outer end of the main shaft (1-4) by means of a coupling (3); A worm wheel (1-5), wherein the worm wheel (1-5) is sleeved on the main shaft (1-4) and is drivingly connected to the worm (1-3); A control circuit board (1-8) is arranged inside the housing (1-1) and fixed on the rear end cover (1-7); the control circuit board (1-8) is electrically connected to the stepping motor (1-2) and the photoelectric switch (1-9).
2. The automatic tracking and monitoring system for spreaders based on AI recognition according to claim 1 is characterized in that: The camera device (2) comprises a movement (2-3), wherein the movement (2-3) is arranged in an outer tube (2-2), and the outer wall of the outer tube (2-2) is connected to a coupling (3); one end of the outer tube (2-2) is connected to a back cover (2-1), and the other end of the outer tube (2-2) is provided with a tempered glass (2-5).
3. The automatic tracking and monitoring system for spreaders based on AI recognition according to claim 2 is characterized in that: The outer cover of the tempered glass (2-5) is provided with a light shield (2-4), and the light shield (2-4) is connected to the end of the outer cylinder (2-2).