Intelligent safety monitoring device for constructional engineering

Abstract

The invention discloses an intelligent safety monitoring device for constructional engineering, and relates to the technical field of safety monitoring. According to the scheme, the system comprises a monitoring part which is used for obtaining the face data of a worker who wants to enter a construction site, inputting the face data into a trained safety identification model, and outputting a safety category; the barrier gate is used for limiting personnel from entering and exiting the construction site; the wearing part is used for being worn on a lower jaw belt of the safety helmet and used for obtaining a pressure value given by the lower jaw belt to the wearing part, the wearing part is connected with the monitoring part through wireless data, and the monitoring part is used for comparing the pressure value with a preset pressure threshold value and judging whether a safety warning instruction is generated or not. Through the cooperation of the wearing part and the monitoring part, once the pressure value monitored by the wearing part is abnormal, the monitoring part can immediately find the abnormal pressure value and inform a safety officer, so that the production safety of a construction site is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring, and in particular to an intelligent safety monitoring device for construction engineering. Background Art

[0002] Safety management of construction sites is particularly important. Wearing hard hats is the most basic safety measure. At present, the safety management of construction sites is basically supervised by safety officers. Safety officers can only patrol or check the wearing of hard hats when entering the construction site, but they cannot monitor the wearing of hard hats by construction site personnel at all times.

[0003] In order to solve the above problems, currently we can use robot-based supervision, such as using machine vision to supervise construction site personnel. However, monitoring through machine vision or inspection robots requires the deployment of a large number of cameras or inspection robots, which is costly, and the construction site is complex and the deployment is also difficult. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose an intelligent safety monitoring device for construction engineering, so as to realize real-time monitoring of the wearing conditions of safety helmets of construction site personnel through the cooperation between the wearing part and the monitoring part.

[0005] In order to achieve the above technical objectives, the present invention provides an intelligent safety monitoring device for construction engineering:

[0006] It includes:

[0007] The monitoring unit is used to obtain the facial data of people who want to enter the construction site, input the facial data into the trained security recognition model, and output the security category;

[0008] Barriers, used to restrict access to and from construction sites;

[0009] The wearing part is used to be worn on the chin strap of the safety helmet and to obtain the pressure value given to the wearing part by the chin strap. The wearing part is connected to the monitoring part via wireless data. The monitoring part is used to compare the pressure value with a preset pressure value threshold to determine whether to generate a safety warning instruction.

[0010] Preferably, the monitoring unit includes a collection unit, a safety identification unit, an analysis unit and a warning unit, and each unit is connected via a wired and / or wireless network;

[0011] The collection unit is used to collect facial data of personnel who intend to enter the construction site;

[0012] The security recognition unit is used to input facial data into the trained security recognition model and output a security category;

[0013] The analysis unit is used to determine whether a qualified instruction is generated based on the safety category, and if a qualified instruction is generated, the gate is opened based on the qualified instruction;

[0014] The warning unit is used to compare the pressure value with the preset pressure value threshold to determine whether to generate a safety warning instruction. If a safety warning instruction is generated, a warning message is sent to the communication terminal specified by the safety officer.

[0015] Preferably, the training method of the security identification model is:

[0016] Pre-collect g groups of historical facial data, where g is an integer greater than 1, and the historical facial data includes image data of correctly wearing a helmet, abnormally wearing a helmet, and not wearing a helmet;

[0017] Numerical annotation is performed on k groups of historical facial data, and the facial data corresponding to the correct wearing of a helmet is marked as 1, and the facial data corresponding to the abnormal wearing of a helmet and the absence of a helmet is marked as 0; each group of historical facial data is used as input of a machine learning model, and the machine learning model uses the predicted annotation of each group of historical facial data as output, the actual annotation as the prediction target, and minimizing the sum of the prediction accuracies of all annotations as the training target; the machine learning model is trained until the sum of the prediction accuracies converges, and the training is stopped and the machine learning model obtained by the training is used as a polishing state discrimination model, and the polishing state discrimination model is a naive Bayes model or a support vector machine model.

[0018] Preferably, the method for the analysis unit to determine whether to generate a qualified instruction includes: if the output mark is 1, a qualified instruction is generated; if the output mark is 0, no qualified instruction is generated.

[0019] Preferably, the barrier is rotatably connected to the outer surface of the monitoring unit, and a limiter is provided on the side of the barrier, the limiter is fixed in the monitoring unit, the limiter is used to limit the barrier, and the limiter is used to retract based on qualified instructions.

[0020] Preferably, the wearing portion includes a first wearing block, a second wearing block and a pressure collection module, the first wearing block and the second wearing block are plugged and fixed, and a surplus space is reserved between the first wearing block and the second wearing block for the pressure collection module to collect pressure values.

[0021] Preferably, the method for the warning unit to determine whether to generate a safety warning instruction includes: if the pressure value is greater than or equal to a preset pressure value threshold, no safety warning instruction is generated; if the pressure value is less than the preset pressure value threshold, a safety warning instruction is generated.

[0022] Preferably, a buckle is fixed to one end of the first wearing block, and the other end of the first wearing block is rotatably connected to a first clamping plate.

[0023] Preferably, a clamping groove is provided at one end of the second wearing block, and a second clamping plate is rotatably connected to the other end of the second wearing block.

[0024] Preferably, the outer surfaces of the first clamping plate and the second clamping plate are evenly fixed with latch teeth.

[0025] It can be seen from the above technical solutions that the present application has the following beneficial effects:

[0026] The monitoring robot is used as a barrier to control the entry of personnel to the construction site. That is, the barrier is used to block personnel from entering the construction site, and they must wear a safety helmet correctly before entering. In addition, through the cooperation between the wearing part and the monitoring part, once the pressure value monitored by the wearing part is abnormal, the monitoring part can immediately discover and notify the safety officer, thereby greatly improving the production safety of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0028] Figure 1 A schematic diagram of the overall structure of an intelligent safety monitoring device for construction engineering provided by the present invention;

[0029] Figure 2 A schematic diagram of the overall structure of a wearing part of an intelligent safety monitoring device for construction engineering provided by the present invention;

[0030] Figure 3 A schematic diagram of the explosion structure of the wearing part of an intelligent safety monitoring device for construction engineering provided by the present invention;

[0031] Figure 4 A schematic diagram of the structure of a safety identification unit of an intelligent safety monitoring device for construction engineering provided by the present invention;

[0032] Figure 5 A schematic diagram of the structure of a warning unit of an intelligent safety monitoring device for construction engineering provided by the present invention.

[0033] Description of the drawings: 1. Monitoring unit; 2. Barrier; 21. Limiter; 3. Wearing unit; 31. First wearing block; 311. Buckle; 312. First splint; 32. Second wearing block; 321. Slot; 322. Second splint. DETAILED DESCRIPTION

[0034] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, and use. It should be understood that in all of these figures, the same or similar reference numerals indicate the same or similar parts and features. The various drawings only schematically represent the concepts and principles of the embodiments of the present disclosure, and do not necessarily show the specific dimensions and proportions of the various embodiments of the present disclosure. Specific parts in specific drawings may be exaggerated to illustrate the relevant details or structures of the embodiments of the present disclosure.

[0035] Embodiment 1

[0036] See also Figure 1 and Figure 4 As shown, an intelligent safety monitoring device for construction engineering includes a monitoring part 1, a barrier 2 and a wearing part 3. The barrier 2 is rotatably connected to the monitoring part and is used to restrict people from entering and exiting the construction site. If a qualified instruction is generated, the barrier 2 is opened based on the qualified instruction; the wearing part 3 is used to be worn on a safety helmet to play a real-time supervision role; it is worth mentioning that the wearing part 3 includes but is not limited to monitoring the pressure value, and the detection part 1 determines whether the safety helmet falls off according to the change of the pressure value. This embodiment adopts the method of determining whether the safety helmet is worn correctly in real time, taking the pressure value determination as an example.

[0037] The monitoring unit 1 includes a collection unit, a safety identification unit, an analysis unit and a warning unit, and each unit is connected through a wired and / or wireless network. The collection unit is used to collect facial data of personnel who want to enter the construction site. In this embodiment, the collection unit adopts a camera, which is installed on the surface of the monitoring unit;

[0038] The security recognition unit is used to input facial data into the trained security recognition model and output the security category; the training method of the security recognition model is:

[0039] In the experimental environment, g groups of historical facial data are pre-collected, where g is an integer greater than 1. The historical facial data includes image data of correctly wearing a helmet, abnormally wearing a helmet, and not wearing a helmet;

[0040] Numerical annotation is performed on k groups of historical facial data, and the facial data corresponding to the correct wearing of a helmet is marked as 1, and the facial data corresponding to the abnormal wearing of a helmet or not wearing a helmet is marked as 0;

[0041] Each set of historical facial data is used as the input of the machine learning model. The machine learning model uses the predicted annotations of each set of historical facial data as output, the actual annotations as the prediction target, and the minimization of the sum of the prediction accuracy of all annotations as the training goal. The calculation formula for the prediction accuracy is: k =(a k -w k ) 2, where k is the number of historical facial data, z k is the prediction accuracy, a k is the predicted annotation value corresponding to the kth group of historical facial data, w k is the actual annotation corresponding to the kth group of historical facial data; the machine learning model is trained until the sum of the prediction accuracies reaches convergence. It should be noted that the convergence standard is determined by those skilled in the art according to the actual situation, the training is stopped and the machine learning model obtained by the training is used as the polishing state discrimination model, and the polishing state discrimination model is a naive Bayes model or a support vector machine model;

[0042] The analysis unit is used to determine whether to generate a qualified instruction based on the safety category. The specific method is: if the output is marked as 1, it means that the wearing is standard, and a qualified instruction is generated; if the output is marked as 0, it means that the wearing is not standard, and no qualified instruction is generated.

[0043] The warning unit is used to compare the pressure value with the preset pressure value threshold to determine whether to generate a safety warning instruction. If a safety warning instruction is generated, a warning message is sent to the communication terminal specified by the safety officer.

[0044] Exemplarily, this embodiment is installed at the entrance of a construction site or a building site. Personnel who want to enter the construction site or the building site need to wear a safety helmet correctly. The monitoring unit 1 detects whether the helmet is worn properly. The facial data is collected and analyzed by the safety recognition model. Before the detection, the wearing part 3 needs to be worn on the chin strap of the safety helmet. The wearing part 3 obtains the pressure value of the chin strap given to the wearing part 3 in real time. The wearing part 3 is connected to the monitoring unit 1 through wireless data. The monitoring unit 1 is used to compare the pressure value with the preset pressure value threshold to determine whether to generate a safety warning instruction. If the user takes off the safety helmet, or the safety helmet falls off and collides during operation, resulting in a change in the pressure value, a safety warning instruction will be generated at this time;

[0045] The purpose is that by monitoring the pressure value through the wearing part 3, it is possible to effectively supervise whether personnel are wearing safety helmets in real time, and in the event of danger or inadvertent fall-off of the safety helmet, it can also be discovered in time by the monitoring part 1 and informed to the safety officer to avoid further occurrence of dangerous accidents.

[0046] For details, see Figure 1 As shown, the barrier 2 is rotatably connected to the outer surface of the monitoring part 1, and a limiter 21 is provided on the side of the barrier 2, and the limiter 21 is fixed in the monitoring part 1. The limiter 21 is used to limit the barrier 2. The limiter 21 is used to retract based on a qualified instruction. The limiter in this embodiment can adopt an electric telescopic rod, a hydraulic cylinder or an electromagnet, which is not specifically limited here. Taking the electric telescopic rod as an example, the electric telescopic rod is extended, and its output end can stop the barrier 2 to achieve the limiting effect.

[0047] See also Figure 2 and Figure 3 As shown, the wearing part 3 includes a first wearing block 31, a second wearing block 32 and a pressure collection module, the first wearing block 31 and the second wearing block 32 are plugged and fixed, and there is a surplus space between the first wearing block 31 and the second wearing block 32 for the pressure collection module to collect pressure values; that is, after wearing, the first wearing block 31 and the second wearing block 32 can slide in a small range, thereby giving pressure to the pressure collection module. The pressure collection module of this embodiment adopts a pressure sensor or a sensor with a pressure detection function, which is not specifically limited here;

[0048] It is worth mentioning that in this embodiment, the first wearing block 31 or the second wearing block 32 is equipped with a battery and a wireless communication module, which are used for power supply and communication respectively. The specific circuit principle will not be elaborated in detail here.

[0049] Specifically, the method for the warning unit to determine whether to generate a safety warning instruction includes: if the pressure value is greater than or equal to a preset pressure value threshold, no safety warning instruction is generated; if the pressure value is less than the preset pressure value threshold, a safety warning instruction is generated; it is worth mentioning that the preset pressure value threshold is obtained by technical personnel in this field based on a large amount of experimental data and is not specifically limited here.

[0050] For further information, see Figure 2 and Figure 3 As shown, a buckle 311 is fixed to one end of the first wearing block 31, and the other end of the first wearing block 31 is rotatably connected to the first clamping plate 312, a slot 321 is opened at one end of the second wearing block 32, and the other end of the second wearing block 32 is rotatably connected to the second clamping plate 322; illustratively, when this embodiment is worn, the first clamping plate 312 and the second clamping plate 322 can be respectively clamped at the two ends of the chin strap, and then the buckle 311 can be engaged with the slot 321 to complete the wearing.

[0051] Furthermore, the outer surfaces of the first clamping plate 312 and the second clamping plate 322 are evenly fixed with the locking teeth to improve the connection stability with the strap.

[0052] Embodiment 2

[0053] An intelligent safety monitoring device for construction projects, based on the first embodiment, provides a magnetic wearing method, replaces the connection structure of the buckle 311 and the slot 321, and a positioning module can be built into the wearing part 3. When a safety warning instruction is generated, the monitoring part 1 can obtain the location of the person concerned in real time, allowing the safety officer to find the person concerned in the shortest time.

[0054] The exemplary implementation scheme of the present disclosure is described in detail above with reference to the preferred embodiments. However, it can be understood by those skilled in the art that, without departing from the concept of the present disclosure, various modifications and variations can be made to the above-mentioned specific embodiments, and various technical features and structures proposed in the present disclosure can be combined in various ways without exceeding the protection scope of the present disclosure, which is determined by the attached claims.

Claims

1. An intelligent safety monitoring device for construction engineering, characterized in that: include: The monitoring unit (1) is used to obtain facial data of personnel who intend to enter the construction site, input the facial data into a trained safety recognition model, and output a safety category; A barrier gate (2) is used to restrict people from entering and leaving the construction site; The wearing part (3) is used to be worn on the chin strap of the helmet and is used to obtain the pressure value applied to the wearing part (3) by the chin strap. The wearing part (3) is connected to the monitoring part (1) via wireless data. The monitoring part (1) is used to compare the pressure value with a preset pressure value threshold to determine whether to generate a safety warning instruction.

2. The intelligent safety monitoring device for construction engineering according to claim 1, characterized in that: The monitoring unit (1) includes a collection unit, a safety identification unit, an analysis unit and a warning unit, and each unit is connected via a wired and / or wireless network; The collection unit is used to collect facial data of personnel who intend to enter the construction site; The security recognition unit is used to input facial data into the trained security recognition model and output a security category; The analysis unit is used to determine whether a qualified instruction is generated based on the security category, and if a qualified instruction is generated, the gate (2) is opened based on the qualified instruction; The warning unit is used to compare the pressure value with the preset pressure value threshold to determine whether to generate a safety warning instruction. If a safety warning instruction is generated, a warning message is sent to the communication terminal specified by the safety officer.

3. The intelligent safety monitoring device for construction engineering according to claim 1, characterized in that: The training method of the security recognition model is: Pre-collect g groups of historical facial data, where g is an integer greater than 1, and the historical facial data includes image data of correctly wearing a helmet, abnormally wearing a helmet, and not wearing a helmet; Numerical annotation is performed on k groups of historical facial data, and the facial data corresponding to the correct wearing of a helmet is annotated as 1, and the facial data corresponding to the abnormal wearing of a helmet and the not wearing of a helmet is annotated as 0; each group of historical facial data is used as input of a machine learning model, and the machine learning model uses the predicted annotation of each group of historical facial data as output, the actual annotation as the prediction target, and minimizing the sum of the prediction accuracies of all annotations as the training target; The machine learning model is trained until the sum of the prediction accuracies reaches convergence, the training is stopped and the trained machine learning model is used as a polishing state discrimination model, and the polishing state discrimination model is a naive Bayes model or a support vector machine model.

4. The intelligent safety monitoring device for construction engineering according to claim 3, characterized in that: The method for the analysis unit to determine whether to generate a qualified instruction includes: if the output mark is 1, a qualified instruction is generated; if the output mark is 0, no qualified instruction is generated.

5. The intelligent safety monitoring device for construction engineering according to claim 2, characterized in that: The barrier gate (2) is rotatably connected to the outer surface of the monitoring part (1), and a limiter (21) is provided on the side of the barrier gate (2). The limiter (21) is fixed in the monitoring part (1), and is used to limit the barrier gate (2). The limiter (21) is used to retract based on a qualified instruction.

6. The intelligent safety monitoring device for construction engineering according to claim 1, characterized in that: The wearing part (3) comprises a first wearing block (31), a second wearing block (32) and a pressure collection module; the first wearing block (31) and the second wearing block (32) are plugged and fixed, and a surplus space is reserved between the first wearing block (31) and the second wearing block (32) for the pressure collection module to collect pressure values.

7. The intelligent safety monitoring device for construction engineering according to claim 1, characterized in that: The method for the warning unit to determine whether to generate a safety warning instruction includes: if the pressure value is greater than or equal to a preset pressure value threshold, no safety warning instruction is generated; if the pressure value is less than the preset pressure value threshold, a safety warning instruction is generated.

8. The intelligent safety monitoring device for construction engineering according to claim 6, characterized in that: A buckle (311) is fixed to one end of the first wearing block (31), and a first clamping plate (312) is rotatably connected to the other end of the first wearing block (31).

9. The intelligent safety monitoring device for construction engineering according to claim 8, characterized in that: One end of the second wearing block (32) is provided with a clamping groove (321), and the other end of the second wearing block (32) is rotatably connected to a second clamping plate (322).

10. The intelligent safety monitoring device for construction engineering according to claim 9, characterized in that: The outer surfaces of the first clamping plate (312) and the second clamping plate (322) are both evenly fixed with latch teeth.

Images