CN216102527U

CN216102527U - Real-time safety helmet identification device based on YOLOV4

Info

Publication number: CN216102527U
Application number: CN202122776843.9U
Authority: CN
Inventors: 孟婷; 杜云刚; 王毅钊
Original assignee: Inner Mongolia University of Technology
Current assignee: Inner Mongolia University of Technology
Priority date: 2021-11-12
Filing date: 2021-11-12
Publication date: 2022-03-22
Anticipated expiration: 2031-11-12

Abstract

The utility model relates to the technical field of safety helmet identification devices, in particular to a real-time safety helmet identification device based on YOLOV4, which comprises an equipment shell, wherein an LCD display screen is arranged in the equipment shell; the LCD screen mounting structure comprises a mounting bracket, a camera and a buffer sponge, wherein the mounting bracket is fixed above the LCD screen and below the LCD screen, the camera is installed above the mounting bracket, the buffer sponge is arranged on the other side of the mounting bracket, the other side of the buffer sponge is connected with an anti-collision telescopic rod, and an expansion spring is sleeved outside the anti-collision telescopic rod. This real-time safety helmet recognition device based on YOLOV4 can realize real time monitoring function, can detect user's helmet wearing condition through the camera that sets up to the action of wearing the helmet is analyzed and is compared through the YOLOV4 algorithm in the PLC controller, and starts through PLC controller control electric vehicle controller, and the user is when not wearing the helmet, and the electric motor car can't normally start, thereby forces the user to wear safety helmet necessarily.

Description

Real-time safety helmet identification device based on YOLOV4

Technical Field

The utility model relates to the technical field of safety helmet identification devices, in particular to a real-time safety helmet identification device based on YOLOV 4.

Background

At present, the intelligent helmet is based on an Arduino single-chip microcomputer, and on the basis, three functions of collision detection, collision alarm and real-time detection are achieved. The product mainly has three functions: firstly, detecting whether a product user collides in the driving process; secondly, the temperature, the pulse and the heart rate of the user can be detected on the mobile phone APP in real time; thirdly, when a product user is in an accident, the helmet is connected with a mobile phone through Bluetooth, the short message reminds an emergency contact person, and body detection data and GPS positioning of a driver are acquired on the mobile phone APP. The helmet is designed with a stretchable flexible liner inside, which is attached to a series of pads on the outside. These pads consist of a proprietary mixture of viscoelastic and non-newtonian fluid materials that become temporarily more viscous (and thus harder) when subjected to pressure, and by virtue of their rheological properties have long-lasting utility in real life, but are perceived by the rider as worn by the helmet. Nowadays, many cities have begun to strictly check traffic violations where non-motor vehicles such as electric vehicles and motorcycles do not wear safety helmets. However, the law enforcement is performed on the public security site or the detection is performed through a street camera, the detection method cannot remind the rider of wearing the helmet in time, a large amount of manpower and funds are consumed, and under the condition that the law enforcement is not strict, a lot of people cannot actively wear the safety helmet or pick up the safety helmet for driving after the detection is finished.

The conventional safety helmet identification device is usually arranged in a helmet, so that whether the helmet is worn during riding cannot be enforced, and a user can take off the helmet after checking.

SUMMERY OF THE UTILITY MODEL

The utility model aims to provide a real-time safety helmet identification device based on YOLOV4, and aims to solve the problems that the safety helmet identification device is usually arranged in a helmet, so that whether the helmet is worn during riding cannot be enforced, and a user can take off the helmet after inspection.

In order to achieve the purpose, the utility model provides the following technical scheme: real-time safety helmet identification device based on YOLOV4 includes:

the device comprises a device shell, wherein an LCD display screen is arranged inside the device shell;

the mounting bracket is fixed above and below the LCD screen, a camera is mounted above the mounting bracket, a buffer sponge is arranged on the other side of the mounting bracket, an anti-collision telescopic rod is connected to the other side of the buffer sponge, and a telescopic spring is sleeved outside the anti-collision telescopic rod;

the loudspeaker is arranged on the right side inside the equipment shell through a screw, and a heat dissipation plate is arranged on one side of the loudspeaker;

and the PLC is installed below the loudspeaker through a screw, and the output end of the PLC is connected with an electric vehicle controller.

Preferably, the inside of buffering sponge is equipped with a plurality of micropore, and the buffering sponge passes through anticollision telescopic link, expanding spring and equipment housing elastic connection to expanding spring is along the surperficial equidistance evenly distributed of buffering sponge.

Preferably, the PLC controller is electrically connected with the camera and the electric vehicle controller, and the PLC controller is electrically connected with the loudspeaker.

Preferably, the anti-collision telescopic rod penetrates through the inside of the telescopic spring and is of a telescopic structure.

Preferably, the device housing is further provided with:

the equipment base is fixed at the rear end of the equipment shell, and the rear end of the equipment base is connected with a connecting rod;

the installed part, it is connected the end of connecting rod, the end of installed part is equipped with the connecting plate, and the inside of connecting plate is worn to be equipped with fixing bolt, pressure sensor is installed to the inboard of connecting plate.

Preferably, the equipment base, the connecting rod and the mounting piece are in pin joint with each other, and the mounting piece is of a circular arc structure.

Preferably, pressure sensor and PLC controller electric connection, the installed part passes through fixing bolt, connecting plate and handlebar constitution detachable construction.

Compared with the prior art, the utility model provides a real-time safety helmet identification device based on YOLOV4, which has the following beneficial effects: this real-time safety helmet recognition device based on YOLOV4 can realize real time monitoring function, can detect user's helmet wearing condition through the camera that sets up to the action of wearing the helmet is analyzed and is compared through the YOLOV4 algorithm in the PLC controller, and starts through PLC controller control electric vehicle controller, and the user is when not wearing the helmet, and the electric motor car can't normally start, thereby forces the user to wear safety helmet necessarily.

1. According to the utility model, the wearing condition of the helmet of the user can be detected through the arranged camera, the monitored data are transmitted to the PLC, and the PLC analyzes and compares the received data, so that the electric vehicle controller is controlled to work, the electric vehicle can be prevented from being started by the user without wearing a safety helmet, and the safety rate of people going out is further improved;

2. according to the utility model, the pressure sensor is arranged, so that the pressure between the handlebar and the connecting plate can be monitored in real time, the monitored data is transmitted to the PLC, the PLC analyzes the received pressure value, and when the pressure value is suddenly changed, the PLC prompts a person to detach the identification equipment without accident through the loudspeaker, so that the warning effect is improved;

3. according to the utility model, through the arrangement of the buffer sponge, the anti-collision telescopic rod and the telescopic spring, the external force received by the equipment shell can be buffered, so that the equipment shell is prevented from directly extruding internal electric elements inwards when collision occurs, and the internal electric elements are protected; through the installed part that sets up can install identification equipment on the electric motor car handlebar fast to identification equipment is with low costs, easy extensive the implementation.

Drawings

FIG. 1 is a schematic view of the overall structure of the present invention;

FIG. 2 is a rear view of the present invention;

FIG. 3 is a schematic view of the internal cross-sectional structure of the housing of the apparatus of the present invention;

fig. 4 is a schematic view of the working process of the present invention.

In the figure: 1. an LCD display screen; 2. a camera; 3. an equipment housing; 4. a connecting rod; 5. a mounting member; 6. fixing the bolt; 7. an equipment base; 8. a connecting plate; 9. a pressure sensor; 10. a heat dissipation plate; 11. a buffer sponge; 12. mounting a bracket; 13. an anti-collision telescopic rod; 14. a PLC controller; 15. a tension spring; 16. a loudspeaker.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1 and 2, the real-time safety helmet identification device based on YOLOV4 includes: the equipment base 7 is fixed at the rear end of the equipment shell 3, and the rear end of the equipment base 7 is connected with the connecting rod 4; the mounting piece 5 is connected to the tail end of the connecting rod 4, the equipment base 7, the connecting rod 4 and the mounting piece 5 are in pin joint with each other, the mounting piece 5 is of an arc-shaped structure, the identification equipment can be quickly mounted on a handlebar of the electric vehicle through the mounting piece 5, the identification equipment is low in cost and easy to push on a large scale, the connecting plate 8 is arranged at the tail end of the mounting piece 5, the fixing bolt 6 penetrates through the connecting plate 8, the pressure sensor 9 is mounted on the inner side of the connecting plate 8, the pressure sensor 9 is electrically connected with the PLC 14, the mounting piece 5 forms a detachable structure with the handlebar through the fixing bolt 6 and the connecting plate 8, the pressure between the handlebar and the connecting plate 8 can be monitored in real time through the pressure sensor 9, the monitored data are transmitted to the PLC 14, and the PLC 14 analyzes the received pressure value, when the pressure value changes suddenly, the PLC 14 prompts a person to detach the identification device without reason through the loudspeaker 16, and the warning effect is improved.

As shown in fig. 3 and 4, the real-time safety helmet identification device based on YOLOV4 includes: the device comprises a device shell 3, wherein an LCD display screen 1 is arranged inside the device shell 3; a mounting bracket 12 which is fixed on the upper and lower sides of the LCD display screen 1, and a camera 2 is arranged above the mounting bracket 12, the wearing condition of the helmet of the user can be detected through the arranged camera 2, the action of wearing the helmet is analyzed and compared through a Yolov4 algorithm in the PLC 14, and the PLC 14 controls the electric vehicle controller to start, when the user does not wear the helmet, the electric vehicle can not start normally, thereby forcing the user to wear the safety helmet, the other side of the mounting bracket 12 is provided with the buffering sponge 11, and the other side of the buffer sponge 11 is connected with an anti-collision telescopic rod 13, the outer part of the anti-collision telescopic rod 13 is sleeved with a telescopic spring 15, a plurality of micropores are arranged inside the buffer sponge 11, the buffer sponge 11 is elastically connected with the equipment shell 3 through an anti-collision telescopic rod 13 and telescopic springs 15, and the telescopic springs 15 are uniformly distributed along the surface of the buffer sponge 11 at equal intervals; the anti-collision telescopic rod 13 penetrates through the inside of the telescopic spring 15, the anti-collision telescopic rod 13 is of a telescopic structure, external force received by the equipment shell 3 can be buffered through the arranged buffering sponge 11, the anti-collision telescopic rod 13 and the telescopic spring 15, so that the equipment shell 3 is prevented from directly extruding internal electric elements inwards when collision occurs, the internal electric elements are protected, the loudspeaker 16 is installed on the right side of the inside of the equipment shell 3 through screws, and a heat dissipation plate 10 is arranged on one side of the loudspeaker 16; PLC controller 14, it passes through the below of mounting screw at speaker 16, PLC controller 14's output is connected with electric vehicle controller, PLC controller 14 and the equal electric connection of camera 2 and electric vehicle controller, and PLC controller 14 and speaker 16 electric connection, camera 2 through setting up can detect the helmet wearing condition of user, and give PLC controller 14 with the data transmission who monitors, PLC controller 14 carries out the analysis and compares received data, thereby control electric vehicle controller carries out work, avoid the user also can start the electric motor car at not wearing safety helmet, and then improve the security rate of people's trip.

The working principle is as follows: when the real-time safety helmet identification device based on the YOLOV4 is used, firstly, the fixing bolt 6 on the connecting plate 8 is screwed to fix the mounting piece 5 sleeved outside the handlebar, the equipment shell 3 is rotated with the connecting rod 4 through the equipment base 7, and the camera 2 faces the electric vehicle seat; secondly, establishing a helmet wearing data set at the background, firstly acquiring image data with higher resolution from an OpenCV development library, marking an image format file by adopting a labelImg tool, marking out the persons without helmets and the persons with helmets in the image to form a corresponding configuration file, and adopting pictures with different angles and different patterns, wherein the number of the pictures is at least 1000 in the process of training the data set, so that the characteristic value extracted from the trained data set is obvious, and the recognition accuracy is improved; then, the database information is recorded on the PLC 14, the camera 2 can collect image information of a user, the PLC 14 is connected with a brake power-off line of the electric vehicle in series, when the situation that a rider wears the helmet is detected, the electric vehicle controller controls the coil to be powered on, the coil is attracted, and a main power supply is switched on; when the helmet is not detected to be worn by the rider, the PLC 14 transmits a negative 5V voltage to the brake signal line, so that the electric vehicle controller controls the electric vehicle to be powered off, and the personal safety of the rider is effectively protected; finally, pressure sensor 9 can carry out real-time supervision to the handlebar and the pressure between connecting plate 8 to the steerable speaker 16 of PLC controller 14 carries out work, and when equipment shell 3 received the external force extrusion, expanding spring 15 and anticollision telescopic link 13 inwards stretch out and draw back, produce outside elasticity, offset external force, thereby play certain cushioning effect.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the utility model, the scope of which is defined in the appended claims and their equivalents.

Claims

1. Real-time safety helmet identification device based on YOLOV4, characterized by including:

the device comprises a device shell (3), wherein an LCD display screen (1) is arranged inside the device shell (3);

the mounting bracket (12) is fixed on the upper side and the lower side of the LCD display screen (1), the camera (2) is mounted above the mounting bracket (12), the other side of the mounting bracket (12) is provided with a buffer sponge (11), the other side of the buffer sponge (11) is connected with an anti-collision telescopic rod (13), and the outer part of the anti-collision telescopic rod (13) is sleeved with a telescopic spring (15);

a loudspeaker (16) which is installed on the right side inside the equipment shell (3) through a screw, and a heat dissipation plate (10) is arranged on one side of the loudspeaker (16);

and the PLC (14) is installed below the loudspeaker (16) through a screw, and the output end of the PLC (14) is connected with an electric vehicle controller.

2. The real-time safety helmet identification device based on YOLOV4 as claimed in claim 1, wherein the inside of the buffer sponge (11) is provided with a plurality of micropores, the buffer sponge (11) is elastically connected with the equipment shell (3) through the anti-collision telescopic rod (13) and the telescopic springs (15), and the telescopic springs (15) are uniformly distributed along the surface of the buffer sponge (11) at equal intervals.

3. The YOLOV 4-based real-time safety helmet identification device according to claim 1, wherein the PLC controller (14) is electrically connected to the camera (2) and the electric vehicle controller, and the PLC controller (14) is electrically connected to the speaker (16).

4. The YOLOV 4-based real-time safety helmet identification device according to claim 1, wherein the anti-collision telescopic rod (13) penetrates inside the telescopic spring (15), and the anti-collision telescopic rod (13) is of a telescopic structure.

5. YOLOV 4-based real-time safety helmet identification device according to claim 1, wherein the equipment housing (3) is further provided with:

the equipment base (7) is fixed at the rear end of the equipment shell (3), and the rear end of the equipment base (7) is connected with a connecting rod (4);

installed part (5), it is connected the end of connecting rod (4), the end of installed part (5) is equipped with connecting plate (8), and the inside of connecting plate (8) is worn to be equipped with fixing bolt (6), pressure sensor (9) are installed to the inboard of connecting plate (8).

6. The YOLOV 4-based real-time safety helmet identification device according to claim 5, wherein the equipment base (7), the connecting rod (4), and the mounting member (5) are pinned to each other, and the mounting member (5) has a circular arc structure inside.

7. The YOLOV 4-based real-time safety helmet identification device according to claim 5, wherein the pressure sensor (9) is electrically connected to the PLC controller (14), and the mounting member (5) forms a detachable structure with the handlebar through the fixing bolt (6) and the connecting plate (8).