KR102344107B1 - Worker safety accident prevention prediction system and method - Google Patents

Abstract

The present invention relates to a system and method for managing and controlling access reflecting the status of workers, which not only perform simple access control but also prevent industrial accidents in consideration of the current status of workers, which capture images for facial recognition, transmit the images to a central server (2), and apply a facial recognition algorithm, using big data and artificial intelligence, to the images to authenticate workers, determine whether the workers are accessible, and determine whether to permit the entry of confirmed workers, and which allow the central server (2) to use the statistics on the heart rate, biological rhythm, movement status, and the like, transmitted from a user terminal (4) worn by a worker, to detect an abnormal status (for example, physical health status, and stress) of the worker and to automatically notify a manager of the abnormal status for preventing industrial accidents. More particularly, the present invention relates to a system and method for preventing and predicting worker accidents. The system comprises: an access controller (1) including an opening and closing means (11), an audio output unit (12), a biorhythm sensor (14), a communication means (15), and a display (16); a central server (2); a thermal imaging camera (3); a user terminal (4); a manager terminal (5); and a database (6). The central server (2) determines whether the worker is accessible based on the biorhythm, stress index, and current physical condition of the worker, and on whether the worker is authenticated, transmits an opening and closing control signal and a display control signal of the opening and closing means (11) to the access controller (1), and transmits the determination result to the manager terminal (5).

Description

WORKER SAFETY ACCIDENT PREVENTION PREDICTION SYSTEM AND METHOD

The present invention relates to a system and method for predicting worker safety accidents.

Recently, as news of deaths due to overwork of logistics workers and casualties related to industrial accidents increase, improvement of the working environment for workers is recognized as a social problem.

The cause of this problem is that the manager cannot understand the status of the workers entering the workplace until they directly report it to the manager. Even if the report is reported to the manager, the manager's subjective judgment determines the reporter's handling, so even if the report is done properly, there may be cases in which proper results cannot be drawn.

In addition, since the general access control system aims to protect human life, property, and information by restricting unnecessary access to important areas within the building, and to provide a safe activity space, continuous interviews or managers are required to detect abnormal situations of individual workers. It had to be observed on a regular basis in person, and it is impossible to prevent occupational accidents caused by the condition of individual workers.

Therefore, it is required to introduce a system that can prevent accidents by accurately identifying the current status of each individual, rather than an access control system for simply controlling the access of workers or temporary workers.

Korean Patent Publication No. 10-2009-0065211 Korean Patent Publication No. 10-2018-0054325

An object of the present invention is to provide a system and method for predicting worker safety accidents that can prevent industrial accidents in consideration of the current status of workers as well as simple access control to solve the above problems.

In addition, after photographing an image for facial recognition, it is transmitted to the central server (2) and applied to a facial recognition algorithm using big data and artificial intelligence to determine whether the worker can be authenticated and workers are allowed to enter, and access the confirmed workers Its purpose is to decide whether or not to allow it.

In addition, the central server 2 uses the statistics on the heart rate, circadian rhythm, movement status, etc. transmitted from the user terminal 4 worn by the worker to the worker's abnormal state (eg, physical health status, stress, etc.). ) is detected in advance and automatically notified to the manager to prevent industrial accidents in advance.

In order to achieve the above object, the present invention provides an access control device (1) including an opening/closing means (11), an audio output unit (12), a biorhythm sensor (14), a communication means (15), and a display (16) ; central server (2); thermal imaging camera (3); user terminal (4); manager terminal (5); and a database (6); wherein the central server (2) determines whether the worker enters and exits the worker based on the worker's biorhythmic stress index, the current physical condition, and whether the worker is authenticated, thereby controlling the opening/closing control signal of the opening/closing means (11). and transmits the display control signal to the access controller (1), and provides a system and method for predicting worker safety accidents, characterized in that the determination result is transmitted to the manager terminal (5).

In addition, whether the authentication of the present invention is determined by the central server 2 using the recognition result of the biorhythm sensor 14, the recognition result of the user terminal 4, and the captured image of the thermal imaging camera 3 characterized.

In addition, the factors for calculating the worker's circadian rhythm stress index of the present invention are at least one of statistical working hours, alcohol consumption, rest time, otolith time, heart rate, pulse, steps, body heat, sleep time, movement, or current location. It is characterized in that it is calculated by combining at least three or more factors among measurable items and the expression of workers.

In addition, the factors for calculating the worker circadian rhythm stress index of the present invention are received from the user terminal 4, the access control unit 1, and the thermal imaging camera 3 and stored in the database 6 do it with

In addition, the factors for calculating the worker's circadian rhythm stress index of the present invention are characterized in that they are calculated by comparing the data stored in the database (6) with the updated data of the worker's current state.

In addition, the central server (2) of the present invention is characterized in that it transmits the calculated current state of the worker and the worker's circadian rhythm stress index to the manager terminal (5).

In addition, the central server (2) of the present invention is characterized in that it transmits the calculated current state of the worker and the worker's status predicted from the worker's circadian rhythm stress index to the manager terminal (5).

According to the present invention, the following effects can be achieved.

The system and method for predicting worker safety accidents according to the present invention as described above prevent industrial accidents by considering the current status of workers (eg, stress, health status, fever, etc.) as well as simple access control. It is possible to reduce human and material damage that may occur as a result.

In addition, since an individual is identified by photographing an image for facial recognition, there is an effect that can solve a security problem that may occur due to theft or misuse of an existing ID.

1 is a view for explaining a system and method for predicting worker safety accidents according to an embodiment of the present invention.
2 is a photographed image of the thermal imaging camera 3 according to an embodiment of the present invention.
3 is a photographed image of the thermal imaging camera 3 according to another embodiment of the present invention.

In describing the present invention, if it is determined that a detailed description of a related well-known function or component may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Elements having substantially the same functional configuration in the drawings are given the same reference numbers and reference numerals as much as possible even though they are shown in different drawings. For convenience of explanation, if necessary, the device and method will be described together.

The suffixes “module” and “unit” for components used in the following description are simply given in consideration of the ease of writing this specification, and the “module” and “unit” may be used interchangeably.

Further, an embodiment of the present invention will be described in detail below with reference to the accompanying drawings and the contents described in the accompanying drawings, but the present invention is not limited or limited by the embodiments.

Hereinafter, in order to clarify the technical spirit of the present invention, a system and method for predicting worker safety accident prevention according to the present invention will be described in detail with reference to the accompanying drawings.

Referring to Figure 1, "worker safety accident prevention prediction system reflecting the worker's condition" according to the present invention is an access controller (1), a central server (2), a thermal imaging camera (3), a user terminal (4), a manager It includes a terminal 5 and a database 6 , and each component is connected through the Internet 7 .

Each of the above-described components drives the access control system by performing the following operations.

[Access Control]

The access controller 1 includes an opening/closing means 11 , an audio output unit 12 , a biorhythm sensor 14 , a communication means 15 , and a display 16 .

The opening/closing means 11 is opened and allowed to enter when an authenticated worker or worker passes, and when an unauthorized worker or worker passes, it maintains a closed state to maintain a secure state.

The opening/closing means 11 may be implemented in various forms, such as a left-right slide type, a bar rotation type, and a vertical blocking type, depending on the installation location or purpose.

The voice output unit 12 outputs a comment related to whether the opening/closing means 11 is operated under the control of the central server 2 or the manager terminal 5, a manager setting guide message according to the situation, an alarm sound, and the like. The output voice is recorded in a separate storage means included in the access controller 1 and the recorded voice data is output in response to a control signal transmitted from the central server 2 or the manager terminal 5 .

The alcohol sensor 13 determines whether a worker or a worker who wants to pass the access control device 1 is drinking.

The operation principle of the alcohol detection sensor 13 is as follows.

When the heating coil inside the alcohol detection sensor 13 is heated, the heating coil is changed to a semiconductor, and when alcohol molecules in the air stick to the heating coil, electrons are generated, which lowers the resistance of the sensor and increases the amount of current. do.

According to the absolute value or change in the amount of current, it is determined whether or not drinking is done, and the determined information is used as basic information to determine whether or not workers enter or not.

The biorhythm sensor 14 is a device for authenticating the worker by recognizing the worker's fingerprint or iris.

The information collected from the biorhythm sensor 14 is compared with the worker information stored in the database 6 of the central server 2 to authenticate the worker.

The communication means 15 performs at least one function of wired LAN, Wi-Fi, Bluetooth, ZigBee, or PLC to communicate with the user terminal, the user terminal 4, or the central server 2 .

The display 16 may display a photographed image of the thermal imaging camera 3 , a detected temperature of a photographing target, data transmitted from the central server 2 , and the like.

[Thermal imaging camera]

The thermal imaging camera (3) detects the distribution of heat in the image taken by the camera, and displays a marker on the image that distinguishes and displays objects with a temperature higher than the temperature set by the worker, and sends a signal to the manager terminal (5), access control device (1), or transmitted to the central server (2).

In addition, the captured image of the worker is transmitted to the central server 2 and used to identify the worker.

When the subject is an object or an animal, the thermal imaging camera measures body heat by using the principle of emitting light of the corresponding wavelength according to the temperature of the body (object).

Since the body temperature of healthy people is between 36 and 37 °C, a measurement of 37.5 °C is high fever, and the limit of fever caused by a virus is 39.5 °C. used

Although the skin hardly reflects external energy heat, the heat dissipation of the human body can be viewed as the same value as the internal temperature, and the temperature can be measured using the wavelength generated from the internal body temperature.

The measured body temperature is used as basic information to determine whether or not to have an infectious disease accompanied by fever and to determine the current status of workers.

Although not shown in the drawings, the thermal imaging camera 3 is configured to include an image analysis unit. The image analyzer may determine an object from the image received from the camera lens, recognize the shape of the object, and extract features.

The feature value of an image means a characteristic that stands out by being distinguished from other parts such as points, borders, edges, textures, and colors that exist in the image.

At least one of a boundary line detection method using a mask, a straight line detection method using Hough transform, and a Harris corner point detection method may be applied as an algorithm for extracting image features.

The image analyzer may extract the features of the object by applying an algorithm suitable for the shape of the object determined from the captured image.

It is necessary to determine whether the object in the photographed image is a person or an object such as a robot vacuum cleaner, but if it is determined that the object in the photographed image is a movement of an organism according to the signal of the infrared sensor by interworking with the infrared sensor, the characteristic value of the object in the photographed image is extracted. An image based on the temperature of the subject is transmitted to the access controller (1), the manager terminal (5), or the central server (2).

[User terminal]

The user terminal 4 may have a form similar in function or appearance to a smart watch in general, and means a terminal capable of accessing the Internet 7 .

The user terminal 4 may be any type of wearable device, and is worn on the body of the worker, such as a smart watch, wearable clothing, safety boots, hard hat, Google Glass, etc. to measure the worker's biometric information and location information and transmit it to the central server 2 . Wearable devices are divided into accessory types (smartwatches, bands, glasses, necklaces), textile clothing integrated types (shirts, socks, shoes), body attachment types (patches, headsets, hats, headbands), and bioimplantable types (pills, chip implantation in body parts). For example, a smart watch may provide biometric information such as heart rate, electrocardiogram, wearable clothing, body temperature, safety shoes, weight, hard hat, brain waves, and Google Glass, vision (indirect resolution measurement). In addition, the user terminal 4 detects whether it is worn on the worker's body and transmits whether it is worn or not to the central server 2 . Whether worn or not may be detected from motion information measured by the user terminal 4 .

In addition, the user terminal 4 may include an RFID or perform an NFC function to authenticate the worker on behalf of the worker's ID.

The user terminal (4) transmits information to the central server (2) by monitoring the current state of the worker, which can be measured, including the wearing worker's pulse, steps, body temperature, sleep time, movement, or current location.

The user terminal 4 may measure the worker's circadian rhythm stress index by using the measured worker information, and transmit the measured worker circadian rhythm stress index to the central server 2 . In addition, the user terminal 4 may display the measured worker circadian rhythm stress index in color and digital values.

The worker's biorhythm stress index may be measured by the user terminal 4 or calculated based on the information transmitted from the user terminal 4 by the central server 2 .

Each user terminal 4 is assigned an ID, so that the manager can manage the worker's information if there is the worker's consent.

[Administrator Terminal]

The manager terminal 5 may be connected to the central server 2 and display data transmitted from the central server 2 .

The manager may perform appropriate operations, such as notifying the upper department based on the data transmitted from the central server 2, if necessary.

In addition, by sending a management control command to the central server (2), it is possible to change the set value for the access controller (1) operation policy, and if necessary, by receiving the stored data from the database (6) connected to the central server (2) can be processed

[Central Server]

The central server 2 is configured to include the database 6, and can determine whether to enter or not by using big data and learning-based artificial intelligence (AI) stored in the database 6, and control access according to the determination result. have.

In addition, the central server (2) is physically or logically connected to the manager terminal (5), the user terminal, the thermal imaging camera (3), and the access controller (1) to perform under the control or command of the manager or to perform a preset command. access control, worker authentication, and worker status check by

The authentication data transmitted from the biorhythm sensor 14 or worker ID information from the user terminal 4 is compared with the worker information stored in the database 6 to authenticate the worker.

In the case of a worker who is permitted access as a result of the authentication, the access controller 1 is instructed to open the opening/closing means 11 to permit access, and the result is transmitted to the access controller 1 and displayed on the display 16 .

In addition, the log related to the worker is recorded in the database 6 along with the entry and exit, and this is notified to the manager.

In the case of a worker whose access is not permitted as a result of authentication, the access control device (1) keeps the opening/closing means (11) closed to disallow entry, and the result is transmitted to the access control device (1) and displayed on the display (16) do.

In addition, the log related to the worker (e.g., the worker's face or the worker's current status, etc.) is recorded in the database 6 along with the access block, and it is sent to the manager. notify

In another embodiment of the present invention, the determination unit of the central server 2 may use the captured image transmitted from the thermal imaging camera 3 to authenticate the worker and determine the condition of the worker.

First, by using the captured image transmitted from the thermal imaging camera (3) to recognize the face and the extracted worker information is compared with the worker information stored in the database (6) to determine whether matching.

If there is matching data in the database 6, the worker is authenticated as an accessible worker and access is allowed.

When the worker uses the user terminal 4 that includes an ID card or RFID or includes an NFC function, the result is recorded in the database 6 by combining it with the access record and the worker's current status information.

In addition, in the case of a certified worker, it can be used as a factor to determine the worker's current state by recognizing the worker's expression from the captured image.

If the determination unit of the central server 2 fails to recognize the face, or if there is no mask wearing or matching data, it is notified to the manager terminal 5 to notify the data currentization or transmit an alarm message.

The log stored in the database 6 includes worker access time, authentication-related data, data that can identify the worker's status, and the like.

Statistical working hours, alcohol consumption, rest time, otolith time, heart rate, pulse, steps, body temperature, sleep time, movement, or current location may be used as data to be stored in the database 6 to determine the state of the worker. can

[Usage of Big Data and Artificial Intelligence]

Although not shown in the drawings, the central server 2 may separately include a storage means for executing an artificial intelligence program, a calculation unit, and a determination unit.

The central server 2 may use the log of workers permitted to enter and exit stored in the database 6 for the repetitive learning function of artificial intelligence.

In general, within a relatively certain range and within a predictable range, workers tend to have a certain pattern of access or movement within the workplace. However, if a worker deviates from the average physical condition due to excessive drinking or overwork the day before entry, it is natural that it will affect the entry time or movement within the workplace.

Although not shown in the drawing, the determination unit of the central server (2) includes the worker's past statistical access time, the worker's current status received from the user terminal (4), the drinking status received from the alcohol sensor (13), and the workplace It is judged whether the worker is in a state suitable for the current job by comprehensively considering the movement time of the worker within the workplace.

As a result of the judgment, if the current status of the worker is not suitable for work, the status of the worker is notified to the manager terminal (5) and the user terminal (4).

In addition, the worker's circadian rhythm stress index may be used to determine whether the worker's current status is suitable for work.

The worker's circadian rhythm stress index includes measurable items including at least one of statistical working hours, alcohol consumption, rest time, otolith time, heart rate, pulse, steps, body temperature, sleep time, movement, or current position, and Comprehensive consideration of the combination of facial expressions, etc., calculates the worker's current state and stress index.

The judgment unit of the central server (2) can detect the worker's abnormal condition in advance by comparing the calculated worker's current status and worker's circadian rhythm stress index with the past worker log. An accident can be prevented in advance by transmitting information on a management target to a user terminal or an administrator terminal.

In addition, it is possible to prevent health problems of workers and serious accidents at the workplace in advance by analyzing the trends of worker status and worker circadian rhythm stress index.

Although the preferred embodiment of the present invention has been described with reference to the components of the access control system reflecting the state of workers, those of ordinary skill in the art will not depart from the spirit and scope of the present invention described in the claims below. It will be understood that various modifications and variations of the present invention can be made without departing from the scope of the present invention.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the following claims as well as components and equivalents thereof.

One … access control device
2 … central server
3 … Thermal imaging camera
4 … user terminal
5 … admin terminal
6 … database
7 … Internet
11 … opening and closing means
12 … audio output unit
13 … alcohol sensor
14 … biorhythm sensor
15 … means of communication
16 … display

Claims (7)

An opening/closing means 11 that can be opened and closed by the control of the central server 2, an audio output unit 12, an alcohol detection sensor 13 including a heating coil whose resistance value is changed by alcohol molecules, and a biological rhythm sensor 14 ), communication means (15), and an access controller (1) including a display (16);
central server (2);
Thermal imaging camera 3 including an infrared sensor and an image analysis unit;
user terminal (4);
manager terminal (5);
and a database (6) in which worker logs are stored; and
When the subject is recognized as a living body in conjunction with the infrared sensor, the image analysis unit recognizes the expression of the worker using a boundary line detection algorithm using a mask, a straight line detection algorithm using Hough transform, and a Harris corner point detection algorithm to recognize the recognized result forward to the central server (2);
The central server 2 determines whether or not the occupant has entered or not based on the entrant's stress index, the entrant's current physical condition, and whether the entrant is authenticated, and transmits the opening/closing control signal and the display control signal of the opening/closing means 11 to the access controller ( 1), and transmits the determination result to the manager terminal 5, and the user terminal 4 and the manager terminal 5 for the state of the person who is predicted from the calculated current physical condition and stress index of the person and to determine the abnormal state of the worker from the difference between the statistics of the worker log including the worker's entry pattern and the worker's movement pattern stored in the database 6 and the current worker's access and movement pattern;
The stress index was analyzed using statistical working hours, alcohol consumption, rest time, otolith time, heart rate, pulse, steps, body heat, sleep time, movement, current location, and recognition results received from the image analysis unit. A worker safety accident prevention prediction system, characterized in that it is calculated by combining the expressions of the entrances and exits. According to claim 1,
The authentication is determined by the central server (2) using the recognition result of the biorhythm sensor (14), the recognition result of the user terminal (4), and the image captured by the thermal imaging camera (3) Worker safety accident prevention prediction system. delete According to claim 1,
The factors for calculating the worker's biorhythm stress index are received from the user terminal (4), the access controller (1), and the thermal imaging camera (3) and stored in the database (6). Accident Prevention Prediction System. 5. The method of claim 4,
Elements for calculating the worker's circadian rhythm stress index are workers' safety accident prevention prediction system, characterized in that it is calculated by comparing the data stored in the database (6) with the updated data of the worker's current status. 6. The method of claim 5,
The central server (2) is a worker safety accident prevention prediction system, characterized in that for transmitting the calculated current state of the worker and the worker's biological rhythm stress index to the manager terminal (5). 7. The method of claim 6,
The central server 2 transmits the calculated current state of the worker and the worker's status predicted from the worker's circadian rhythm stress index to the user terminal 4 and the manager terminal 5, characterized in that the worker safety accident Preventive Prediction System.

Images