Human shape tracking monitoring based on ai vision algorithm

Abstract

The invention relates to the technical field of tracking and monitoring, and discloses a human shape tracking and monitoring system based on an ai vision algorithm, which comprises a mounting holder, a steering angle sensor is fixedly mounted on the inner bottom wall of the mounting holder, a motor controller is fixedly connected to the bottom of a microcomputer controller, an infrared lens is fixedly mounted on the front surface of a lens casing, and a camera is fixedly mounted on the front surface of the infrared lens casing. An imaging assembly is arranged in the lens shell, and a sound sensing assembly is arranged at the top of the lens shell. According to the human shape tracking monitoring based on the ai vision algorithm, infrared radiation data of a human body is captured through an infrared lens, imaging is carried out after image processing, the position of a portrait is judged through an algorithm model, an environment sound source is collected through a sound sensor, and the position of human voice is judged through the algorithm model. Therefore, the monitoring imaging system and the sound source detection system cooperatively judge the human shape position, then the microcomputer controller and the motor controller control the stepping motor, and the rotating direction and the rotating angle of the camera are controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of tracking and monitoring, and in particular to human tracking and monitoring based on AI vision algorithms. Background Art

[0002] The monitoring system is one of the most widely used systems in the security system. The most common ones are the surveillance cameras distributed in the streets and alleys, which can monitor the surrounding environment in a timely manner. Monitoring equipment is needed to capture video images before or during dangerous events, so as to effectively prevent the occurrence of dangerous events or to provide important information for event handling.

[0003] After searching, it was found that a tracking monitoring method and a tracking monitoring device were disclosed in the invention patent with the Chinese patent publication number CN105812721A. The tracking monitoring method and the tracking monitoring device in the invention patent collect audio data of the surrounding environment, determine the position of the speaking sound, and adjust the tracking position of the camera on the human body;

[0004] However, the tracking and monitoring method and the tracking and monitoring equipment only rely on sound detection and judgment. When this judgment method is used in a noisy environment, the collected sound data will include sound source data generated by driving, pedestrians, construction work and activities inside buildings. This will increase the difficulty of sound judgment and cannot be combined with infrared portrait recognition technology to coordinate with sound source detection to determine the position of the human body. Therefore, there is still room for further improvement in the tracking and monitoring method. Therefore, human tracking monitoring based on AI visual algorithm is proposed. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the shortcomings of the prior art, the present invention provides human tracking monitoring based on AI visual algorithm, which has the advantage of collaboratively determining the position of a human figure by combining infrared imaging and sound source detection. It solves the problem that the tracking monitoring method and tracking monitoring equipment in the above-mentioned background technology only rely on sound detection and are difficult to accurately determine the position of the human body in a noisy environment.

[0007] (II) Technical solution

[0008] In order to achieve the above-mentioned purpose of determining the position of a human figure by coordinating infrared imaging with sound source detection, the present invention provides the following technical solutions: human figure tracking monitoring based on AI visual algorithm, including a pan-tilt installation and a turntable rotatably connected to the top of the pan-tilt installation, wherein an angle adjustment structure is arranged inside the pan-tilt installation, and a camera device is installed on the top of the pan-tilt installation through the turntable movement;

[0009] The angle adjustment structure includes a positioning frame fixedly connected to the inner right wall of the mounting platform, a stepping motor is fixedly installed on the inner right wall of the positioning frame, a deceleration assembly is arranged on the inner side of the positioning frame, a steering angle sensor is fixedly installed on the inner bottom wall of the mounting platform, a microcomputer controller is fixedly installed on the inner left wall of the mounting platform, and a motor controller is fixedly connected to the bottom of the microcomputer controller;

[0010] The camera device includes a lens housing fixedly connected to the top of the turntable, an infrared lens fixedly installed on the front of the lens housing, an infrared illuminator fixedly installed on the front of the lens housing, an imaging component is arranged inside the lens housing, and a sound sensor component is arranged on the top of the lens housing.

[0011] Preferably, a partition is fixedly connected to the inner side of the positioning frame, the reduction assembly includes two spiral gear plates rotatably connected to the right side of the partition, the output end of the stepper motor is fixedly connected to a coupling, and the left end of the coupling is coaxially fixed to the right side of one of the spiral gear plates.

[0012] Preferably, the left side of the partition is rotatably connected to two first bevel gears, the right ends of the two first bevel gears are coaxially fixed to the left sides of the two spiral gear plates, the inner top wall and the inner bottom wall of the positioning frame are rotatably connected to the second bevel gear, and the outer sides of the two second bevel gears are respectively meshed with the two first bevel gears.

[0013] Preferably, the diameters of the two second bevel gears are three times the diameters of the two first bevel gears, respectively, and the top of one of the second bevel gears is coaxially fixed with the bottom of the turntable, and an optical coupling element and a perforated slot plate are provided inside the steering angle sensor, and the top of the perforated slot plate is fixedly connected with a rotating shaft structure, and the top of the rotating shaft is coaxially fixed with the bottom of another first bevel gear.

[0014] Preferably, the optical coupling element includes a light-emitting diode and a photosensitive crystal transistor structure, the perforated slot plate is arranged in the middle gap between the light-emitting diode and the photosensitive crystal transistor, the steering angle sensor is electrically connected to the microcomputer controller through the photosensitive crystal transistor, a single-chip microcomputer module is arranged inside the microcomputer controller, the motor controller is electrically connected to the single-chip microcomputer module through the control circuit, and the stepper motor is electrically connected to the motor controller.

[0015] Preferably, the infrared illuminator is an infrared lamp structure distributed in a circular array outside the infrared lens, the imaging component includes an infrared sensor arranged inside the infrared lens, the internal electrical control of the lens housing is equipped with a processing chip module, the processing chip module includes an image sensor connected to the infrared sensor signal, the image sensor is electrically connected to a D / A converter through a signal processing circuit, and the D / A converter signal is connected to a digital signal processor.

[0016] Preferably, the processing chip module is externally connected to a monitoring imaging system through a digital signal processor, and the monitoring imaging system includes an image display module, a visual algorithm recognition unit and an image model classification unit, the visual algorithm recognition unit is provided with an AI visual recognition algorithm, and the image model classification unit is provided with a character classification training model.

[0017] Preferably, a cover plate is fixedly connected to the top of the lens housing, the sound sensor assembly includes a positioning plate fixedly connected to the inner top wall of the cover plate, and eight sound sensors are fixedly installed on the bottom of the positioning plate, and the eight sound sensors are distributed in a circular array.

[0018] Preferably, each of the sound sensors is connected to an external sound source detection system via a cable, and the sound source detection system includes a sound collection module, a sound recognition and ranging unit, and a sound source model classification unit. The sound recognition and ranging unit is provided with a sound source recognition algorithm and a sound source ranging algorithm, and the sound source model classification unit is provided with a human voice classification training model.

[0019] (III) Beneficial effects

[0020] Compared with the prior art, the present invention provides a human tracking monitoring based on AI visual algorithm, which has the following beneficial effects:

[0021] The human tracking monitoring based on AI visual algorithm captures the infrared radiation of the human body through an infrared lens to form infrared signal data, which is received by the infrared sensor, converted into an electrical signal and transmitted to the image sensor. The image sensor generates an image data signal, which is then amplified and converted by the D / A converter and transmitted to the digital signal processor. The image is imaged in the image display module of the monitoring imaging system. The AI ​​visual recognition algorithm of the visual algorithm recognition unit recognizes the human image and object image data, and then the human image data is separated by the character classification training model to determine the position of the human image. The sound sensor is distributed in a circular array position, which can evenly collect multi-directional sound sources in the surrounding environment. The sound source recognition algorithm detects human voices and other types of sounds. At the same time, the sound source ranging algorithm is used to detect the distance of the sound source, and the human voice data is separated in the human voice classification training model to determine the position of the human voice, so that the monitoring imaging system and the sound source detection system can collaboratively determine the position of the human figure, and then the turntable is driven by a stepper motor to drive the camera device to rotate the angle, and the steering angle sensor detects the rotation direction and rotation angle in real time, which is convenient for the angle adjustment of the camera device. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A three-dimensional schematic diagram of the monitoring structure of the present invention;

[0023] Figure 2 for Figure 1 Cross-sectional view of perspective;

[0024] Figure 3 for Figure 2 Schematic diagram of the deceleration component structure in the perspective;

[0025] Figure 4 for Figure 2 Schematic diagram of the structure of the sound sensor component in the perspective;

[0026] Figure 5 is a schematic diagram of an imaging assembly of the present invention;

[0027] Figure 6 This is a framework diagram of the monitoring imaging system of the present invention;

[0028] Figure 7 It is a framework diagram of the sound source detection system of the present invention;

[0029] Figure 8 This is a flow chart of the stepper motor control method of the present invention.

[0030] In the figure: 1. Install the pan / tilt; 2. Angle adjustment structure; 201. Positioning frame; 202. Stepping motor; 203. Speed ​​reduction assembly; 2031. Rotary gear plate; 2032. First bevel gear; 2033. Second bevel gear; 204. Steering angle sensor; 205. Microcomputer controller; 206. Motor controller; 3. Camera device; 301. Lens housing; 302. Infrared lens; 303. Infrared illuminator; 304. Imaging assembly; 3041. Infrared sensor; 3042. Image sensor; 3043. D / A converter; 3044. Digital signal processor; 305. Sound sensing assembly; 3051. Positioning plate; 3052. Sound sensor; 4. Coupling; 5. Cover plate. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1-8 , human tracking monitoring based on AI visual algorithm, including a mounting platform 1 and a turntable rotatably connected to the top of the mounting platform 1, an angle adjustment structure 2 is arranged inside the mounting platform 1, and a camera device 3 is installed on the top of the mounting platform 1 through the turntable movement;

[0033] The angle adjustment structure 2 includes a positioning frame 201 fixedly connected to the inner right wall of the mounting platform 1, a stepping motor 202 is fixedly installed on the inner right wall of the positioning frame 201, a deceleration component 203 is arranged on the inner side of the positioning frame 201, a steering angle sensor 204 is fixedly installed on the inner bottom wall of the mounting platform 1, a microcomputer controller 205 is fixedly installed on the inner left wall of the mounting platform 1, and a motor controller 206 is fixedly connected to the bottom of the microcomputer controller 205;

[0034] The camera device 3 includes a lens housing 301 fixedly connected to the top of the turntable, an infrared lens 302 fixedly installed on the front of the lens housing 301, an infrared illuminator 303 fixedly installed on the front of the lens housing 301, an imaging component 304 is arranged inside the lens housing 301, and a sound sensor component 305 is arranged on the top of the lens housing 301.

[0035] Furthermore, a partition is fixedly connected to the inner side of the positioning frame 201, and the reduction assembly 203 includes two spiral gear plates 2031 rotatably connected to the right side of the partition. The output end of the stepper motor 202 is fixedly connected to a coupling 4, and the left end of the coupling 4 is coaxially fixed to the right side of one of the spiral gear plates 2031.

[0036] Specifically, the output end of the stepper motor 202 is used to drive the coupling 4, and the coupling 4 drives one of the spiral toothed plates 2031. The two spiral toothed plates 2031 are meshed with each other for transmission, and the rotation directions of the two spiral toothed plates 2031 are opposite to each other.

[0037] Furthermore, the left side of the partition is rotatably connected to two first bevel gears 2032, and the right ends of the two first bevel gears 2032 are coaxially fixed to the left sides of the two spiral gear plates 2031 respectively. The inner top wall and the inner bottom wall of the positioning frame 201 are rotatably connected to the second bevel gear 2033, and the outer sides of the two second bevel gears 2033 are respectively meshed with the two first bevel gears 2032.

[0038] Specifically, the two spiral toothed plates 2031 are respectively connected to the right sides of the two first bevel gears 2032 to ensure that the two first bevel gears 2032 rotate in opposite directions, and then the outer sides of the left ends of the two first bevel gears 2032 are respectively meshed with the two second bevel gears 2033. Since the two second bevel gears 2033 are symmetrically arranged up and down, and the two first bevel gears 2032 are both located on the right sides of the two second bevel gears 2033, the rotation speed and rotation direction of the two second bevel gears 2033 are the same at this time.

[0039] Furthermore, the diameters of the two second bevel gears 2033 are three times the diameters of the two first bevel gears 2032, respectively. The top of one of the second bevel gears 2033 is coaxially fixed with the bottom of the turntable. The interior of the steering angle sensor 204 is provided with an optical coupling element and a perforated slot plate. The top of the perforated slot plate is fixedly connected with a rotating shaft structure, and the top of the rotating shaft is coaxially fixed with the bottom of the other first bevel gear 2032.

[0040] Specifically, since the diameter of the second bevel gear 2033 is three times the diameter of the first bevel gear 2032, the second bevel gear 2033 is driven by the engagement of the first bevel gear 2032, and then the two second bevel gears 2033 are respectively rotatably connected to the turntable and the steering angle sensor 204, which can reduce the rotation speed of the turntable. At the same time, the steering angle sensor 204 is used to detect the rotation direction and angular speed of the camera device 3.

[0041] Furthermore, the optical coupling element includes a light-emitting diode and a photosensitive crystal transistor structure, the perforated slot plate is arranged in the middle gap between the light-emitting diode and the photosensitive crystal transistor, the steering angle sensor 204 is electrically connected to the microcomputer controller 205 through the photosensitive crystal transistor, the microcomputer controller 205 is internally provided with a single-chip microcomputer module, the motor controller 206 is electrically connected to the single-chip microcomputer module through the control circuit, and the stepper motor 202 is electrically connected to the motor controller 206.

[0042] Specifically, the steering angle sensor 204 is composed of an optical coupling element composed of a light emitting diode and a photosensitive crystal triode and a rotatably arranged perforated slot plate as main components, and the rotating shaft portion at the top of the perforated slot plate is connected to another first bevel gear 2032, and the rotating shaft is driven by the first bevel gear 2032, so that the perforated slot plate rotates in the gap between the light emitting diode and the photosensitive crystal triode;

[0043] A plurality of light-transmitting circular holes are provided on the perforated slot plate, and the light emitted by the light-emitting diode is projected into the photosensitive crystal triode. At this time, the photosensitive crystal triode detects different light signals and generates corresponding data differential signals according to the changes of the light passing through the perforated slot plate, and transmits the data differential signals to the signal processing module of the microcomputer controller 205, and the steering angle, rotation direction and rotation speed of the second bevel gear 2033 and the camera device 3 are determined according to the data signals;

[0044] The motor controller 206 is electrically connected to the stepper motor 202 through a relay module and a pulse control circuit. When the single-chip microcomputer module receives the sound source position signal and the portrait position signal, the motor controller 206 is controlled to output a pulse signal of a certain frequency, and the speed, angle and steering data of the stepper motor 202 are controlled by the pulse signal.

[0045] Furthermore, the infrared illuminator 303 is an infrared lamp structure distributed in a circular array outside the infrared lens 302, the imaging component 304 includes an infrared sensor 3041 arranged inside the infrared lens 302, the internal electrical control of the lens housing 301 is equipped with a processing chip module, the processing chip module includes an image sensor 3042 connected to the signal of the infrared sensor 3041, the image sensor 3042 is electrically connected to a D / A converter 3043 through a signal processing circuit, and the D / A converter 3043 is signal-connected to a digital signal processor 3044.

[0046] Specifically, the infrared lens 302 captures infrared radiation from the human body to form infrared signal data, which is received by the infrared sensor 3041. The infrared sensor 3041 then converts the infrared signal into an electrical signal and transmits it to the image sensor 3042. The image sensor 3042 generates an image data signal, which is then amplified and converted by the D / A converter 3043 and then transmitted to the digital signal processor 3044 for subsequent processing.

[0047] Furthermore, the processing chip module is externally connected to a monitoring imaging system through a digital signal processor 3044. The monitoring imaging system includes an image display module, a visual algorithm recognition unit and an image model classification unit. The visual algorithm recognition unit is provided with an AI visual recognition algorithm, and the image model classification unit is provided with a character classification training model.

[0048] Specifically, the digital signal processor 3044 transmits the converted digital image data to the image display module of the monitoring imaging system for imaging, and the AI ​​visual recognition algorithm of the visual algorithm recognition unit recognizes the portrait and object image data, and then the human classification training model separates the portrait data and determines the portrait position.

[0049] Furthermore, a cover plate 5 is fixedly connected to the top of the lens housing 301, and the sound sensor assembly 305 includes a positioning plate 3051 fixedly connected to the inner top wall of the cover plate 5, and eight sound sensors 3052 are fixedly installed on the bottom of the positioning plate 3051, and the eight sound sensors 3052 are distributed in a circular array.

[0050] Specifically, the sound sensors 3052 are distributed in a circular array and can evenly collect sound sources in the surrounding environment.

[0051] Furthermore, each sound sensor 3052 is connected to an external sound source detection system via a cable. The sound source detection system includes a sound collection module, a sound recognition and ranging unit, and a sound source model classification unit. The sound recognition and ranging unit is provided with a sound source recognition algorithm and a sound source ranging algorithm, and the sound source model classification unit is provided with a human voice classification training model.

[0052] Specifically, the sound sensor 3052 transmits the detected sound signal to the sound collection module, and the sound source recognition algorithm detects the human voice and other types of sounds. At the same time, the sound source ranging algorithm is used to detect the distance of the sound source, and the human voice data is separated in the human voice classification training model to determine the human voice position.

[0053] In summary, the human tracking monitoring based on AI visual algorithm captures the infrared radiation of the human body through the infrared lens 302 to form infrared signal data, which is received by the infrared sensor 3041, and the infrared signal is converted into an electrical signal and transmitted to the image sensor 3042. The image sensor 3042 generates an image data signal, which is then amplified and converted by the D / A converter 3043 and transmitted to the digital signal processor 3044. The image is imaged in the image display module of the monitoring imaging system, and the AI ​​visual recognition algorithm of the visual algorithm recognition unit recognizes the human and object image data, and then the human portrait data is separated by the character classification training model. , determine the position of the portrait, the sound sensor 3052 is distributed in a circular array position, and can evenly collect multi-directional sound sources in the surrounding environment. The sound source recognition algorithm detects human voices and other types of sounds. At the same time, the sound source ranging algorithm is used to detect the sound source distance, and the human voice data is separated in the human voice classification training model to determine the position of the human voice. The monitoring imaging system and the sound source detection system are used to collaboratively determine the position of the human figure, and then the stepper motor 202 drives the turntable to drive the camera device 3 to rotate the angle, and the steering angle sensor 204 detects the rotation direction and rotation angle in real time, which is convenient for the angle adjustment of the camera device 3.

[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0055] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A human tracking monitoring system based on AI visual algorithm, comprising a mounting platform (1) and a turntable rotatably connected to the top of the mounting platform (1), characterized in that: An angle adjustment structure (2) is provided inside the mounting platform (1), and a camera device (3) is movably mounted on the top of the mounting platform (1) via a turntable; The angle adjustment structure (2) comprises a positioning frame (201) fixedly connected to the inner right wall of the mounting platform (1); a stepping motor (202) is fixedly mounted on the inner right wall of the positioning frame (201); a deceleration component (203) is arranged on the inner side of the positioning frame (201); a steering angle sensor (204) is fixedly mounted on the inner bottom wall of the mounting platform (1); a microcomputer controller (205) is fixedly mounted on the inner left wall of the mounting platform (1); and a motor controller (206) is fixedly connected to the bottom of the microcomputer controller (205); The camera device (3) comprises a lens housing (301) fixedly connected to the top of the turntable, an infrared lens (302) fixedly mounted on the front of the lens housing (301), an infrared illuminator (303) fixedly mounted on the front of the lens housing (301), an imaging component (304) arranged inside the lens housing (301), and a sound sensor component (305) arranged on the top of the lens housing (301).

2. The human tracking monitoring based on AI visual algorithm according to claim 1 is characterized by: A partition is fixedly connected to the inner side of the positioning frame (201), the deceleration assembly (203) comprises two spiral toothed discs (2031) rotatably connected to the right side of the partition, the output end of the stepping motor (202) is fixedly connected to a coupling (4), and the left end of the coupling (4) is coaxially fixed to the right side of one of the spiral toothed discs (2031).

3. The human tracking monitoring based on AI visual algorithm according to claim 2 is characterized by: The left side of the partition is rotatably connected to two first bevel gears (2032), the right ends of the two first bevel gears (2032) are coaxially fixed to the left sides of the two spiral toothed discs (2031), the inner top wall and the inner bottom wall of the positioning frame (201) are rotatably connected to second bevel gears (2033), and the outer sides of the two second bevel gears (2033) are respectively meshed with the two first bevel gears (2032).

4. The human tracking monitoring based on AI visual algorithm according to claim 3 is characterized by: The diameters of the two second bevel gears (2033) are three times the diameters of the two first bevel gears (2032), the top of one of the second bevel gears (2033) is coaxially fixed with the bottom of the turntable, an optical coupling element and a perforated slot plate are arranged inside the steering angle sensor (204), a rotating shaft structure is fixedly connected to the top of the perforated slot plate, and the top of the rotating shaft is coaxially fixed with the bottom of another first bevel gear (2032).

5. The human tracking monitoring based on AI visual algorithm according to claim 4 is characterized by: The optical coupling element comprises a light emitting diode and a photosensitive crystal triode structure, the perforated slot plate is arranged at the intermediate gap between the light emitting diode and the photosensitive crystal triode, the steering angle sensor (204) is electrically connected to a microcomputer controller (205) via the photosensitive crystal triode, a single-chip microcomputer module is arranged inside the microcomputer controller (205), the motor controller (206) is electrically connected to the single-chip microcomputer module via a control circuit, and the stepping motor (202) is electrically connected to the motor controller (206).

6. The human tracking monitoring based on AI visual algorithm according to claim 1 is characterized by: The infrared illuminator (303) is an infrared lamp structure distributed in a circular array outside the infrared lens (302); the imaging component (304) comprises an infrared sensor (3041) arranged inside the infrared lens (302); the internal electrical control of the lens housing (301) comprises a processing chip module; the processing chip module comprises an image sensor (3042) signal-connected to the infrared sensor (3041); the image sensor (3042) is electrically connected to a D / A converter (3043) via a signal processing circuit; and the D / A converter (3043) is signal-connected to a digital signal processor (3044).

7. The human tracking monitoring based on AI visual algorithm according to claim 6 is characterized by: The processing chip module is externally connected to a monitoring imaging system via a digital signal processor (3044); the monitoring imaging system comprises an image display module, a visual algorithm recognition unit and an image model classification unit; the visual algorithm recognition unit is provided with an AI visual recognition algorithm, and the image model classification unit is provided with a character classification training model.

8. The human tracking monitoring based on AI visual algorithm according to claim 7 is characterized by: The top of the lens housing (301) is fixedly connected to a cover plate (5), the sound sensor assembly (305) comprises a positioning plate (3051) fixedly connected to the inner top wall of the cover plate (5), and eight sound sensors (3052) are fixedly installed on the bottom of the positioning plate (3051), and the eight sound sensors (3052) are distributed in a circular array.

9. The human tracking monitoring based on AI visual algorithm according to claim 8 is characterized by: Each of the sound sensors (3052) is connected to an external sound source detection system via a cable. The sound source detection system comprises a sound collection module, a sound recognition and ranging unit, and a sound source model classification unit. The sound recognition and ranging unit is provided with a sound source recognition algorithm and a sound source ranging algorithm. The sound source model classification unit is provided with a human voice classification training model.

Images