Construction site safety early warning monitoring system based on deep learning unmanned aerial vehicle

Abstract

The invention relates to the field of monitoring systems, and provides a construction site safety early warning monitoring system based on a deep learning unmanned aerial vehicle, which comprises an unmanned aerial vehicle body, and is characterized in that the unmanned aerial vehicle body is provided with a safety monitoring system and a safety helmet carrying mechanism; the safety monitoring system comprises a movable camera, and a pattern recognition module, a deep learning module and an alarm module are arranged in the movable camera; the safety helmet carrying mechanism comprises two bearing and clamping mechanisms, each bearing and clamping mechanism comprises a rod body, the rod bodies are movably connected to the unmanned aerial vehicle body, two cap supporting strips are movably connected to the rod bodies, and friction plates are connected to the cap supporting strips through second elastic pieces. When the unmanned aerial vehicle recognizes that a person does not wear the safety helmet, the unmanned aerial vehicle body can directly carry the safety helmet to fly to the position of the target person, instant delivery of the safety helmet is achieved, an extra electric driving device is not needed, and the unmanned aerial vehicle has the advantages of being stable and convenient to clamp and release the safety helmet and high in adaptability.

Description

Technical Field

[0001] This invention relates to the field of monitoring system technology, specifically to a construction site safety early warning and monitoring system based on deep learning unmanned aerial vehicles (UAVs). Background Technology

[0002] Chinese invention patent CN119399792B discloses a construction site safety monitoring system based on UAV vision. This invention relates to the field of construction site safety monitoring technology and solves the problem of not being able to uniformly identify whether relevant construction workers are wearing safety helmets and whether their helmets are worn correctly. This invention uses UAVs to monitor the construction area of ​​the construction site in real time and determine the monitoring screen in real time. Then, based on relevant extracted features, it selects facial images from the monitoring screen and performs preliminary analysis on the facial images. By locking the center point and the point to be processed, it identifies whether the mean angle of the connecting lines within the outline is consistent. This can preliminarily determine whether the helmet of the corresponding construction worker is worn abnormally and generate relevant judgment signals for timely display. Its monitoring method is more comprehensive and can achieve better construction site safety supervision.

[0003] The aforementioned patent addresses how to identify individuals not wearing safety helmets, but it cannot promptly provide helmets to these individuals, thus failing to ensure their safety in a timely manner. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention aims to provide a construction site safety early warning and monitoring system based on deep learning-based unmanned aerial vehicles (UAVs). To solve these problems, this invention employs the following technical solution: A construction site safety early warning and monitoring system based on deep learning drones includes a drone body, on which a safety monitoring system and a safety helmet carrying mechanism are installed; The security monitoring system includes a mobile camera, which is equipped with an image recognition module, a deep learning module, and an alarm module. The safety helmet carrying mechanism includes two support and clamping mechanisms. Each support and clamping mechanism includes a rod that is movably connected to the drone body. Two helmet support strips are movably connected to the rod, and friction plates are connected to the helmet support strips via elastic elements.

[0005] Optionally, the rod body includes a fixed rod and a movable rod. The fixed rod is slidably connected to the UAV body, and the movable rod is slidably connected to the fixed rod. The fixed rod is connected to the movable rod through an elastic element. A rack is fixedly connected to the fixed rod, and a component cavity is opened on the movable rod. The rack extends into the component cavity. The cap strip is slidably connected to the inner wall of the component cavity, and extends to the outside of the movable rod. A rack two is fixedly connected to the cap strip. A spur gear is rotatably connected to the inner wall of the component cavity. The rack one and the two rack two are meshed with the spur gear. A permanent magnet and a locking plug are slidably connected to the inner wall of the component cavity. The permanent magnet extends to the bottom of the movable rod. A locking piece is fixedly connected to the permanent magnet. A lock and an airbag two are fixedly connected to the inner wall of the component cavity. The lock has a lock groove. A press-to-lock mechanism is provided in the lock groove. The press-to-lock mechanism and the lock piece are adapted to each other. The airbag two and the locking plug are fixedly connected. On one of the cap strips, a cavity is formed. A trigger strip, a limiting permanent magnet strip, and a pressure plate are slidably connected to the inner wall of the cavity. One end of the trigger strip extends to the outside of the cap strip. A limiting groove is formed on the trigger strip. The trigger strip is connected to the inner wall of the cavity through an elastic element three. The limiting permanent magnet strip and the pressure plate are fixedly connected. The pressure plate is connected to the inner wall of the cavity through an elastic element four. An airbag one is fixedly connected to the inner wall of the cavity. The airbag one is connected to the airbag two through a hose.

[0006] Optionally, the friction pad has a curved surface.

[0007] Optionally, a friction layer is connected to the curved surface.

[0008] Optionally, the friction layer material may include silicone.

[0009] Optionally, the movable camera also includes a noise reduction module and an enhancement module.

[0010] Optionally, the movable camera also includes a data cache module.

[0011] Optionally, the movable camera may also include an anomaly detection module.

[0012] Optionally, the alarm module includes a speaker.

[0013] Optionally, a ground control terminal is also included, which is wirelessly connected to the UAV and has a touch screen display.

[0014] The present invention has the following beneficial effects: This invention incorporates a safety monitoring system on a drone, utilizing a mobile camera to monitor construction sites and employing image recognition and deep learning modules to identify whether personnel are wearing safety helmets. When a helmet is detected, the drone can directly carry it to the target personnel's location, delivering it instantly. This prevents personnel from remaining in a dangerous working state due to not wearing helmets, significantly improving the timeliness of safety assurance at construction sites. It requires no additional electric drive and offers advantages such as stable and convenient helmet clamping, easy release, and strong adaptability. Attached Figure Description

[0015] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of a construction site safety early warning and monitoring system based on deep learning UAVs, according to the present invention. Figure 2 This is a schematic diagram of the structure of the UAV body in this invention; Figure 3 This is a schematic diagram of the structure of the fixed rod and the movable rod in this invention; Figure 4 This is the present invention. Figure 3 Enlarged view of point A in the middle; Figure 5 This is a schematic diagram of an existing safety helmet structure.

[0017] Reference numerals in the attached diagram: 1. UAV body; 2. Fixed rod; 3. Movable rod; 4. Rack 1; 5. Elastic element 1; 6. Component cavity; 7. Bearing strip; 8. Elastic element 2; 9. Friction plate; 10. Rack 2; 11. Spur gear; 12. Trigger strip; 13. Limiting groove; 14. Elastic element 3; 15. Limiting permanent magnet strip; 16. Pressure plate; 17. Elastic element 4; 18. Airbag 1; 19. Hose; 20. Airbag 2; 21. Locking plug; 22. Permanent magnet; 23. Locking plate; 24. Locking device; 25. Locking groove; 26. Strip cavity. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] In the description of this invention, it should be noted that the terms "vertical," "upper," "lower," "horizontal," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0020] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or a connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] like Figures 1-4 As shown, a construction site safety early warning and monitoring system based on deep learning UAVs includes a UAV body 1, on which a safety monitoring system and a safety helmet carrying mechanism are provided; The security monitoring system includes a mobile camera, which is equipped with an image recognition module, a deep learning module, and an alarm module. The safety helmet carrying mechanism includes two support and clamping mechanisms. Each support and clamping mechanism includes a rod that is movably connected to the drone body 1. Two helmet support strips 7 are movably connected to the rod. Friction pieces 9 are connected to the helmet support strips 7 via elastic elements 8. The friction pieces 9 are used to position and clamp the safety helmet.

[0022] Based on the above scheme, in some embodiments, the rod body includes a fixed rod 2 and a movable rod 3. The fixed rod 2 is slidably connected to the UAV body 1, and the movable rod 3 is slidably connected to the fixed rod 2. The fixed rod 2 is connected to the movable rod 3 through an elastic element 5. A rack 4 is fixedly connected to the fixed rod 2, and a component cavity 6 is opened on the movable rod 3. The rack 4 extends into the component cavity 6. The cap strip 7 is slidably connected to the inner wall of the component cavity 6, and extends to the outside of the movable rod 3. A rack 2 10 is fixedly connected to the cap strip 7. A spur gear 11 is rotatably connected to the inner wall of the component cavity 6. The rack 1 4 and the two racks 2 10 are all meshed with the spur gear 11. A permanent magnet 22 and a locking plug 21 are slidably connected to the inner wall of the component cavity 6. The permanent magnet 22 extends to the bottom of the movable rod 3. A locking piece 23 is fixedly connected to the permanent magnet 22. A lock 24 and an airbag 20 are fixedly connected to the inner wall of the component cavity 6. A lock groove 25 is opened on the lock 24. A press-lock mechanism is provided in the lock groove 25. The press-lock mechanism and the locking piece 23 are adapted. The airbag 20 and the locking plug 21 are fixedly connected. When the locking piece 23 is pressed on the press-lock mechanism for the first time, it can be interlocked. Then the locking piece 23 rebounds a short distance. When the locking piece 23 is pressed on the press-lock mechanism again, it can be unlocked. The design can refer to the existing technology.

[0023] On one of the cap-support strips 7, a cavity 26 is formed. A trigger strip 12, a limiting permanent magnet strip 15, and a pressure plate 16 are slidably connected to the inner wall of the cavity 26. One end of the trigger strip 12 extends to the outside of the cap-support strip 7, and a limiting groove 13 is formed on the trigger strip 12. The trigger strip 12 is connected to the inner wall of the cavity 26 through an elastic element 3 14. The limiting permanent magnet strip 15 and the pressure plate 16 are fixedly connected. The pressure plate 16 is connected to the inner wall of the cavity 26 through an elastic element 4 17. An airbag 18 is fixedly connected to the inner wall of the cavity 26. The airbag 18 is connected to the airbag 20 through a hose 19. The airbag 18, hose 19, and airbag 20 are all made of soft material.

[0024] In a preferred embodiment of the invention, the friction pad 9 has a curved surface to fit the shape of the inner wall of the safety helmet.

[0025] To increase friction, a friction layer is connected to the curved surface.

[0026] Furthermore, the friction layer material includes silicone, which has good adhesion and sufficiently high friction.

[0027] In addition, the movable camera also includes a noise reduction module and an enhancement module. The noise reduction module is used to suppress noise in the image data acquired by the movable camera, while the enhancement module is used to enhance the image's sharpness, contrast, or brightness, thereby improving image quality and increasing the recognition accuracy of the image recognition module and deep learning module in complex construction environments.

[0028] Optionally, the mobile camera may also include one or more of a data caching module or an anomaly detection module. The data caching module is used to temporarily store the acquired image data or recognition results for subsequent processing or backtracking; the anomaly detection module is used to judge the working status or recognition process of the mobile camera and to promptly prompt or handle any abnormal situations, thereby improving the stability and reliability of the system operation.

[0029] To provide an auditory alert to personnel, the alarm module includes a speaker.

[0030] In a preferred configuration, the system further includes a ground control terminal, which is wirelessly connected to the UAV body 1. The ground control terminal is equipped with a touch screen display. Users can input commands through the touch screen display to control the operating status of the UAV body 1.

[0031] Implementation process: such as Figure 5 As shown, existing safety helmets have multiple top straps inside, with gaps between adjacent top straps and buffer gaps between the top straps and the inner wall of the safety helmet. The function of the buffer gaps is to provide deformation buffer distance and absorb energy when the safety helmet is impacted.

[0032] Clamping operation for safety helmets: Place the safety helmet upside down on the ground, adjust the distance between the two fixed rods 2 to fit different sizes of safety helmets, suspend the drone body 1 in the air, control the drone body 1 to descend, and insert the two movable rods 3 into the gaps between the two top straps respectively; On one of the movable rods 3, the lower end of the permanent magnet 22 abuts against the inner wall of the safety helmet. The permanent magnet 22 is pushed upward by the inner wall of the safety helmet, and the locking plate 23 is inserted into the locking groove 25 and locked with the pressing self-locking mechanism in the locking groove 25. The permanent magnet 22 generates an upward magnetic repulsion force on the limiting permanent magnet strip 15. The upper end of the limiting permanent magnet strip 15 abuts against the bottom wall of the trigger strip 12. As the drone body 1 descends, the elastic element 5 is gradually compressed, and the rack 4 moves downward, driving the spur gear 11 to rotate. The spur gear 11 drives two The racks 2 and 10 move away from each other, thus causing the two helmet straps 7 to move away from each other. The helmet straps 7 pass through the buffer gap between the top strap and the inner wall of the helmet, supporting the top strap and preventing the helmet from falling off. Then, the two friction pads 9 abut against the inner wall of the helmet. If one of the friction pads 9 abuts against the inner wall of the helmet earlier, it will push the helmet to move slightly, causing the other friction pad 9 to also abut against the inner wall of the helmet. The elastic element 2 and 8 are compressed, making the friction pads 9 fit tightly against the inside of the helmet. The silicone friction layer on the wall increases friction. When one of the friction pads 9 pushes the trigger strip 12 to overcome the elastic force of the elastic element 3 14 and move towards the limiting permanent magnet strip 15, when the limiting groove 13 moves above the limiting permanent magnet strip 15, the limiting permanent magnet strip 15 is inserted into the limiting groove 13 under the magnetic repulsion of the permanent magnet element 22, thereby limiting the trigger strip 12. The pressure plate 16 moves upward to compress the airbag 18. Some of the air in the airbag 18 passes through the hose 19 into the airbag 20, causing the airbag to... The expansion of the second 20 causes the locking plug 21 to move upward and insert into the teeth of the spur gear 11, thereby limiting the spur gear 11. At this time, the rack 4 can no longer move downward, controlling the drone body 1 to rise. Under the action of its own gravity, the safety helmet will move down a small distance. The helmet support strip 7 supports the top strap. The locking plate 23 will move down a small distance under its own gravity but will still remain locked to the self-locking mechanism. The four friction plates 9 clamp and position the safety helmet to prevent violent shaking during flight and affect flight.

[0033] The other movable lever 3 operates on the same principle, thereby ensuring that the other two friction plates 9 also come into close contact with the inner wall of the safety helmet.

[0034] The drone body 1 cruises, and the mobile camera acquires image data. The image recognition module identifies whether the person in the image data is wearing a safety helmet. The recognition principle of the image recognition module can refer to the existing technology CN119399792B. The deep learning module is used to learn the image features of workers wearing safety helmets and not wearing safety helmets to improve the recognition accuracy of the image recognition module. When the image recognition module identifies that a person is not wearing a safety helmet, the speaker on the alarm module will sound a prompt, and then the drone body 1 flies to the person and delivers the safety helmet.

[0035] Safety helmet delivery operation: The drone body 1 descends, and the safety helmet touches the ground; On one of the movable rods 3, the lower end of the permanent magnet 22 abuts against the inner wall of the safety helmet. The permanent magnet 22 and the locking plate 23 move upward a short distance, the drone body 1 rises, and the permanent magnet 22 moves downward under its own gravity. The locking plate 23 releases and presses the self-locking mechanism. After the permanent magnet 22 moves away from the limiting permanent magnet strip 15, the limiting permanent magnet strip 15 loses the magnetic repulsion of the permanent magnet 22. The limiting permanent magnet strip 15 and the pressure plate 16 move downward under the elastic force of the elastic element 4 17. The pressure plate 16 releases the compression on the airbag 1 18, and part of the airbag 2 20... Air flows back into the first airbag 18, the second airbag 20 shrinks, the locking plug 21 moves down to release the limit on the spur gear 11, the limiting permanent magnet strip 15 releases the limit on the trigger strip 12, the movable rod 3 moves down relative to the fixed rod 2 under the elastic force of the elastic element 5, the rack 4 drives the spur gear 11 to reverse, the spur gear 11 drives the two cap strips 7 to move closer to each other, the trigger strip 12 resets under the elastic force of the third elastic element 14, thereby releasing the two friction plates 9 from the inner wall of the safety helmet, the cap strips 7 release from the support of the top strap, and the movable rod 3 can then be removed from the safety helmet; The other movable lever 3 operates on the same principle, allowing the safety helmet to remain on the ground for people to pick up and use.

[0036] The beneficial effects of this invention are: This invention installs a safety monitoring system on the drone body 1, uses a mobile camera to patrol and monitor the construction site, and uses an image recognition module and a deep learning module to identify whether personnel are wearing safety helmets. When a person is detected not wearing a safety helmet, the drone body 1 can directly carry the safety helmet to the target person's location, achieving immediate delivery of the safety helmet. This avoids personnel from being in a dangerous working state due to not wearing a safety helmet, and significantly improves the timeliness of safety assurance at the construction site. Meanwhile, this invention, through the setting of a linkage clamping structure composed of a fixed rod 2, a movable rod 3, a rack 1 4, a rack 2 10, a spur gear 11, and a helmet support strip 7, and in conjunction with friction plates 9, elastic element 2 8, limiting permanent magnet strip 15, permanent magnet 22, locking plug 21, etc., enables the safety helmet to achieve automatic support, reliable clamping, and stable limiting during flight, preventing shaking and falling off. After delivery, the clamping is automatically released by airbag 1 18, airbag 2 20 and related mechanisms. The structure is simple, requires no additional electric drive device, and has the advantages of stable and convenient clamping, convenient release, and strong adaptability.

[0037] The components, modules, mechanisms, and devices in this invention that are not described in detail are all general standard parts or components known to those skilled in the art. Their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A construction site safety early warning and monitoring system based on deep learning-based unmanned aerial vehicles (UAVs), characterized in that, Includes the drone body (1), which is equipped with a safety monitoring system and a safety helmet carrying mechanism; The security monitoring system includes a mobile camera, which is equipped with an image recognition module, a deep learning module, and an alarm module. The safety helmet carrying mechanism includes two support and clamping mechanisms. Each support and clamping mechanism includes a rod that is movably connected to the drone body (1). Two helmet support strips (7) are movably connected to the rod. Friction pieces (9) are connected to the helmet support strips (7) through elastic element two (8).

2. The construction site safety early warning and monitoring system based on deep learning UAVs according to claim 1, characterized in that, The rod body includes a fixed rod (2) and a movable rod (3). The fixed rod (2) is slidably connected to the UAV body (1), and the movable rod (3) is slidably connected to the fixed rod (2). The fixed rod (2) is connected to the movable rod (3) through an elastic element (5). A rack (4) is fixedly connected to the fixed rod (2), and a component cavity (6) is opened on the movable rod (3). The rack (4) extends into the component cavity (6). The cap strip (7) is slidably connected to the inner wall of the component cavity (6), and the cap strip (7) extends to the outside of the movable rod (3). A rack two (10) is fixedly connected to the cap strip (7). A spur gear (11) is rotatably connected to the inner wall of the component cavity (6). The rack one (4) and the two rack two (10) are meshed with the spur gear (11). A permanent magnet (22) and a locking plug (21) are slidably connected to the inner wall of the component cavity (6). The permanent magnet (22) extends to the bottom of the movable rod (3). A locking piece (23) is fixedly connected to the permanent magnet (22). A lock (24) and an airbag two (20) are fixedly connected to the inner wall of the component cavity (6). A lock groove (25) is opened on the lock (24). A self-locking mechanism is provided in the lock groove (25). The self-locking mechanism and the locking piece (23) are adapted to each other. The airbag two (20) and the locking plug (21) are fixedly connected. On one of the cap strips (7), a strip cavity (26) is provided. A trigger strip (12), a limiting permanent magnet strip (15) and a pressure plate (16) are slidably connected to the inner wall of the strip cavity (26). One end of the trigger strip (12) extends to the outside of the cap strip (7). A limiting groove (13) is provided on the trigger strip (12). The trigger strip (12) is connected to the inner wall of the strip cavity (26) through the elastic element three (14). The limiting permanent magnet strip (15) and the pressure plate (16) are fixedly connected. The pressure plate (16) is connected to the inner wall of the strip cavity (26) through the elastic element four (17). An airbag one (18) is fixedly connected to the inner wall of the strip cavity (26). The airbag one (18) is connected to the airbag two (20) through the hose (19).

3. The construction site safety early warning and monitoring system based on deep learning UAVs according to claim 2, characterized in that, The friction plate (9) has a curved surface.

4. The construction site safety early warning and monitoring system based on deep learning UAVs according to claim 3, characterized in that, A friction layer is connected to the curved surface.

5. A construction site safety early warning and monitoring system based on deep learning UAVs according to claim 4, characterized in that, The friction layer material includes silicone.

6. A construction site safety early warning and monitoring system based on deep learning UAVs according to any one of claims 1-5, characterized in that, The movable camera also includes a noise reduction module and an enhancement module.

7. A construction site safety early warning and monitoring system based on deep learning UAVs according to claim 6, characterized in that, The movable camera also has a data cache module.

8. A construction site safety early warning and monitoring system based on deep learning UAVs according to claim 6, characterized in that, The movable camera is also equipped with an anomaly detection module.

9. A construction site safety early warning and monitoring system based on deep learning UAVs according to claim 6, characterized in that, The alarm module includes a speaker.

10. A construction site safety early warning and monitoring system based on deep learning UAVs according to claim 6, characterized in that, It also includes a ground control terminal, which is wirelessly connected to the UAV body (1), and the ground control terminal is equipped with a touch screen display.

Images