KR102500474B1 - System for providing smart intelligence safety management service - Google Patents

Abstract

Provided is a system for providing a smart intelligent safety managing service comprising: at least one inputting terminal; at least one outputting terminal; and a server for providing a safety managing service. The at least one inputting terminal photographs an image by using at least one type of camera, and outputs the image in real time. The at least one outputting terminal outputs a risk event analyzed as the image of the at least one inputting terminal. The server for providing a safety managing service includes: a receiving unit which receives the image received from the at least one inputting terminal; an inputting unit which inputs the image as a query to at least one type of artificial intelligence algorithm; and an outputting unit which transmits a risk event to the at least one outputting terminal when the risk event is outputted from the at least one artificial intelligence algorithm. The present invention can prevent industrial accidents in advance.

Description

Smart intelligent safety management service provision system {SYSTEM FOR PROVIDING SMART INTELLIGENCE SAFETY MANAGEMENT SERVICE}

The present invention relates to a smart intelligent safety management service providing system, and provides a platform for preventing accidents in advance by predicting the movement of workers, materials and equipment and notifying dangerous situations in advance in order to prevent industrial accidents.

According to Korea's Sustainable Development Goals Implementation Report 2021 reported by the National Statistical Office in 2021, South Korea's industrial accident death rate was the fourth highest among Organization for Economic Cooperation and Development (OECD) countries. Although the government has presented a national task to reduce the death toll from industrial accidents, which is close to 1,000 each year, by half, and has focused on preventing and supervising industrial accidents, the number of deaths from industrial accidents in 2020 was 882, an increase from 855 in the previous year. In response, the government introduced the Severe Disaster Punishment Act, which asks business owners or management managers to take more responsibility, and will be enforced from January 27, 2022. Efforts to reduce the occurrence of industrial accidents are being emphasized by both the government and companies. There are two major strategic perspectives for reducing the occurrence of industrial accidents. Back) method and the Feed-Forward method, which predicts the extent to which industrial accidents will occur in the future and establishes a long-term or short-term industrial accident prevention strategy to respond to.

At this time, in order to prevent industrial accidents, a method of monitoring industrial sites using drones, sensors, and cameras and guiding workers and managers immediately when dangers occur has been researched and developed. In this regard, Korea Patent Publication No. 2017-0060973 (published on June 2, 2017) and Korean Registered Patent No. 10-1265564 (published on May 20, 2013), a drone flies along a normal flight path and transmits images captured at an industrial accident management site to a control server. , the control server judges whether the industrial accident management site is in an abnormal situation and guides the manager, and when the worker wears the wearable device, the worker's surroundings are photographed or detected and transmitted to the control server, A configuration for inferring whether a worker is in a safe state with respect to surrounding conditions and providing a control signal to the wearable device according to the inference result is disclosed.

However, in the case of the former, continuous monitoring is not easy because it is affected by weather conditions and it is not easy to maintain an altitude and shoot at a constant angle because it must be filmed with a drone. Even in the latter case, the wearable device can only measure the environment close to the worker, so it is unavoidable to be crushed and killed by heavy machinery at a distance, such as in a crane accident. In actual construction sites and industrial sites, human casualties continue even if work is stopped or stopped due to situations in which construction schedules must be met or delivery deadlines must be met, or the number of people responsible for safety is reduced. Even if personnel are supplemented, mechanical objectivity cannot be maintained and 24-hour monitoring is not possible due to differences in personal inclination and background knowledge of the supplemented personnel, which can lead to waste of labor costs. Therefore, it is required to research and develop a platform that can learn the industrial accident situation using artificial intelligence as a feedback method and predict industrial accidents and prevent them in advance, such as the feed-forward method.

An embodiment of the present invention analyzes images collected from at least one camera in real time to identify and classify objects such as people, equipment, and materials, and in the case of people, to determine whether or not preset safety rules are followed Guidance and warning, collect industrial accident data set to build big data, model at least one artificial intelligence algorithm through learning and testing, input equipment and materials input in real time as a query, and then enter the industry It is possible to provide a smart intelligent safety management service providing system that can prevent the occurrence of industrial accidents in advance by outputting the result of predicting whether or not an accident will occur, warning each worker and reporting it to the manager. However, the technical problem to be achieved by the present embodiment is not limited to the technical problem as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention uses at least one type of camera to photograph and output at least one input terminal in real time, and an image of the at least one input terminal. At least one output terminal outputting the analyzed risk event and a receiving unit receiving an image received from at least one input terminal, an input unit inputting the image as a query to at least one type of artificial intelligence algorithm, and at least one input unit. and a safety management service providing server including an output unit for transmitting a dangerous event to at least one output terminal when a dangerous event is output from the artificial intelligence algorithm of the above.

According to any one of the above-described problem solving means of the present invention, images collected from at least one camera are analyzed in real time to identify and classify objects such as people, equipment, and materials, and in the case of people, a predetermined safety rule is observed. After determining whether or not there is a problem, providing guidance and warning, collecting industrial accident datasets to build big data, modeling at least one artificial intelligence algorithm through learning and testing, and querying images of equipment and materials input in real time. It is possible to prevent the occurrence of industrial accidents in advance by outputting the result of predicting the occurrence of industrial accidents after entering it in (Query), warning each worker and reporting it to the manager.

1 is a diagram for explaining a smart intelligent safety management service providing system according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a safety management service providing server included in the system of FIG. 1 .
3 and 4 are diagrams for explaining an embodiment in which a smart intelligent safety management service is implemented according to an embodiment of the present invention.
5 is an operation flowchart for explaining a smart intelligent safety management service providing method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention with reference to the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, this means that it may further include other components, not excluding other components, unless otherwise stated, and one or more other characteristics. However, it should be understood that it does not preclude the possibility of existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

As used throughout the specification, the terms "about", "substantially", etc., are used at or approximating that value when manufacturing and material tolerances inherent in the stated meaning are given, and do not convey an understanding of the present invention. Accurate or absolute figures are used to help prevent exploitation by unscrupulous infringers of the disclosed disclosure. The term "step of (doing)" or "step of" as used throughout the specification of the present invention does not mean "step for".

In this specification, a "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, and two or more units may be realized by one hardware. On the other hand, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or configured to reproduce one or more processors. Thus, as an example, '~unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Functions provided within components and '~units' may be combined into smaller numbers of components and '~units' or further separated into additional components and '~units'. In addition, components and '~units' may be implemented to play one or more CPUs in a device or a secure multimedia card.

In this specification, some of the operations or functions described as being performed by a terminal, device, or device may be performed instead by a server connected to the terminal, device, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed by a terminal, apparatus, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal mean mapping or matching the terminal's unique number or personal identification information, which is the terminal's identifying data. can be interpreted as

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram for explaining a smart intelligent safety management service providing system according to an embodiment of the present invention. Referring to FIG. 1 , the smart intelligent safety management service providing system 1 may include at least one input terminal 100, a safety management service providing server 300, and at least one output terminal 400. However, since the smart intelligent safety management service providing system 1 of FIG. 1 is only one embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 .

At this time, each component of FIG. 1 is generally connected through a network (Network, 200). For example, as shown in FIG. 1 , at least one input terminal 100 may be connected to the safety management service providing server 300 through the network 200 . In addition, the safety management service providing server 300 may be connected to at least one input terminal 100 and at least one output terminal 400 through the network 200 . In addition, at least one output terminal 400 may be connected to the safety management service providing server 300 through the network 200 .

Here, the network refers to a connection structure capable of exchanging information between nodes such as a plurality of terminals and servers, and examples of such networks include a local area network (LAN) and a wide area network (WAN: Wide Area Network), the Internet (WWW: World Wide Web), wired and wireless data communications networks, telephone networks, and wired and wireless television communications networks. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi , Internet (Internet), LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth (Bluetooth) network, NFC ( A Near-Field Communication (Near-Field Communication) network, a satellite broadcasting network, an analog broadcasting network, a Digital Multimedia Broadcasting (DMB) network, etc. are included, but not limited thereto.

In the following, the term at least one is defined as a term including singular and plural, and even if at least one term does not exist, each component may exist in singular or plural, and may mean singular or plural. It will be self-evident. In addition, the singular or plural number of each component may be changed according to embodiments.

At least one input terminal 100 may be a terminal for taking pictures of a construction site, a manufacturing site, etc. indoors and outdoors using a smart intelligent safety management service related web page, app page, program or application. At this time, the input terminal 100 may be any device capable of taking pictures, and for example, various devices such as smart glasses, CCTVs, action cams, and body cams may be used. However, it is not limited to those listed and is not excluded for reasons not listed.

Here, at least one input terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop, a desktop, a laptop, and the like equipped with a navigation system and a web browser. In this case, at least one input terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one input terminal 100 is, for example, a wireless communication device that ensures portability and mobility, and includes navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet ) may include all types of handheld-based wireless communication devices such as terminals, smartphones, smart pads, tablet PCs, and the like.

The safety management service providing server 300 may be a server that provides a smart intelligent safety management service web page, app page, program or application. In addition, the safety management service providing server 300 collects the industrial accident data set, learns at least one artificial intelligence algorithm, and warns in advance when an image input in real time is similar to the data set to prevent accidents. It may be a server that allows In addition, the safety management service providing server 300 may be a server that measures the moving direction, size, speed, etc. of people, materials and equipment, predicts the moving direction and collision, and outputs the result to the output terminal 400 . In addition, the safety management service providing server 300 may be a server that identifies persons, materials, and equipment violating safety laws and regulations and transmits a warning to the output terminal 400 . In addition, the safety management service providing server 300 may be a server that measures or predicts a risk level and reports it to an administrator terminal (not shown). In addition, the safety management service providing server 300 may be a server that accumulates the number of violations and risk levels (levels) in the same workplace and imposes a preset penalty when the number exceeds a preset number or reports to related agencies and ministries.

Here, the safety management service providing server 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop, a desktop, a laptop, and the like equipped with a navigation system and a web browser.

At least one output terminal 400 may be a terminal that outputs prediction, measurement, analysis, and guidance results using a smart intelligent safety management service-related web page, app page, program, or application. In this case, the at least one output terminal 400 may be a monitor, a speaker, a smartphone as a user terminal, a warning light, a laser indicator light, etc., but is not limited to those listed.

Here, at least one output terminal 400 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop, a desktop, a laptop, and the like equipped with a navigation system and a web browser. In this case, at least one output terminal 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one output terminal 400 is, for example, a wireless communication device that ensures portability and mobility, such as navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet ) may include all types of handheld-based wireless communication devices such as terminals, smartphones, smart pads, tablet PCs, and the like.

2 is a block diagram for explaining a safety management service providing server included in the system of FIG. 1, and FIGS. 3 and 4 show an embodiment in which a smart intelligent safety management service is implemented according to an embodiment of the present invention. It is a drawing for explanation.

Referring to FIG. 2, the safety management service providing server 300 includes a receiving unit 310, an input unit 320, an output unit 330, a blind spot removal unit 340, a display unit 350, and a prediction unit 360. , An environment detection unit 370 and a violation monitoring unit 380 may be included.

The safety management service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one input terminal 100 and at least one output terminal 400 is smart intelligent safety management When a service application, program, app page, web page, etc. is transmitted, at least one input terminal 100 and at least one output terminal 400 are smart intelligent safety management service applications, programs, app pages, web pages, etc. can be installed or opened. Also, a service program may be driven in at least one input terminal 100 and at least one output terminal 400 by using a script executed in a web browser. Here, the web browser is a program that allows users to use the web (WWW: World Wide Web) service, and means a program that receives and displays hypertext described in HTML (Hyper Text Mark-up Language). For example, Netscape , Explorer, Chrome, and the like. In addition, an application means an application on a terminal, and includes, for example, an app running on a mobile terminal (smart phone).

Referring to FIG. 2 , the receiving unit 310 may receive an image received from at least one input terminal 100 . At least one input terminal 100 may capture an image using at least one type of camera and output the image in real time. Recently, the Severe Disaster Punishment Act has been enacted and enforced, but punishment does not mean that a person who died or suffered an accident in a disaster can return or go back to before being injured. Since the most important thing is to prevent accidents before people get hurt, a method of observing each material, equipment, facility, etc., predicting an industrial accident before the material breaks or falls, or getting caught, and then notifying the worker immediately. to be received and analyzed in real time for use.

The input unit 320 may input an image as a query to at least one type of artificial intelligence algorithm. The artificial intelligence algorithm is an algorithm that is learned, tested, and modeled with image data sets such as falls, jams, bumps, kicks, overturns, hit by objects, and traffic accidents that occurred at construction or manufacturing sites across the country as input. Accordingly, when a similar image is input, a warning may be given. In addition, when an object suddenly collapses, it is possible to predict the movement path or movement of materials or equipment by inclination, horizontal axis or vertical axis, or movement speed. make it possible

<Personal protective equipment detection>

Personal Protective Equipment (PPE) detection uses computer vision and AI technology to determine whether or not workers are wearing protective equipment. PPE includes hard hats, safety vests, seat belts, safety shoes, and gloves. In Korea, it is mandatory for employers to provide protective equipment and for workers to wear protective equipment in accordance with the Industrial Safety and Health Standards Regulations and the Industrial Safety and Health Act. Site workers often work without wearing protective equipment that is stuffy and obstructive, and site managers have limitations in monitoring all workers. Therefore, if the PPE detection technology automatically checks whether workers are wearing protective equipment, it can protect workers' safety, reduce human and time efforts in supervision, and increase work efficiency.

The core technology applied to PPE detection is object detection technology. Object detection is an automated technique for identifying an object of interest in an image by distinguishing it from the background. For object detection, a deep learning algorithm may be applied, and for example, a You Only Look Once (YOLO) algorithm may be used. YOLO shows fast processing speed and relatively high detection accuracy compared to other deep learning-based object detection algorithms. Because of this, YOLO is used in various real-time object detection applications including PPE detection, but there is a chronic problem of YOLO. That is, it is a weak part in small object detection. In one embodiment of the present invention, YOLOv4 to which various techniques such as Bag of Freebies and Bag of Specials are applied may be used to improve detection performance.

CCTV cameras used for construction site monitoring and PPE detection are installed at various site locations. Workers look relatively small in the video output of cameras installed at a distance to secure a wide viewing angle in an outdoor environment. When performing YOLOv4-based PPE detection on images from cameras installed in such locations, protective equipment that appears small is often not detected. Therefore, in one embodiment of the present invention, when detecting PPE in a video taken from a distance, an image preprocessing technique capable of improving detection performance for a small-sized PPE object may be further used. In YOLOv4, the size of the image input for object detection is resized to be the same as the size of the input image set during learning. Therefore, even if a high-resolution image is provided as an input, a phenomenon in which the size of the original input image is reduced or adjusted occurs, which affects object detection.

Detection performance can be improved if an input image maintaining the quality of the original image is provided through a preprocessing process. Accordingly, when an object is detected in an image frame, only a meaningful part is extracted and provided as an input. That is, only the parts where motion occurs in the video frame are extracted, combined and reconstructed in a mosaic manner, and provided as input to the YOLOv4 object detection model. By extracting only the region of interest or a meaningful part, the size of the input image can be reduced and the original resolution of the region to be detected in the image can be maintained as much as possible. Through this, the detection performance for small objects can be improved.

<object tracking>

In order to apply the image-based location tracking method at the construction site, the 3D location of construction resources such as vehicles and personnel can be successfully tracked using only cameras, but there is no automated entity matching process for each entity, so many It is difficult to track objects simultaneously. Due to the nature of the image-based location tracking method using stereo images, it is necessary to detect and track an object in a captured image, as well as to pair the same object among objects detected in two images. If this process is not automated, it is virtually impossible to track the location of multiple objects using images. Accordingly, in one embodiment of the present invention, an image-based location tracking method capable of tracking multiple objects may be used by adding an object matching process based on epipolar geometry. In the entity matching process, an epipolar line that can be calculated from a fundamental matrix is used as entity matching information.

<Location tracking of multiple personnel>

The location tracking method of multiple construction workers can complement the camera calibration process and object matching process. A separate object matching process can be added to track the location of multiple objects, and camera calibration can be performed using GCP instead of a checkerboard. To this end, calculation of the base matrix from the first frame of each image is preceded. The calculated basis matrix is used not only in the camera calibration process but also in the object matching process. In each image, objects are detected and tracked independently. The tracked entities perform an entity matching process using the calculated basis matrix, and finally calculate 3D position coordinates using these results.

<Object Matching>

For entity matching, epipolar geometry can be applied. The epipolar line is calculated using the elementary matrix from the object center point obtained in the object tracking process, and the object matching is determined by comparing the distance between the calculated epipolar line and the center point of the tracked object. To calculate the basis matrix, feature points are extracted using the first frame of each image. There are various feature point extraction algorithms, but the most commonly used SIFT (Scale Invariant Feature Fransform) algorithm and SURF (Speeded Up Robust Features) algorithm can be compared to use an algorithm that returns more feature points. Feature points extracted from the left and right images are compared, and only the same corresponding points are used for base matrix calculation, and MSAC (M-estimate Sample Consensus) can be used to determine the corresponding points in this process.

<3D position coordinate calculation>

In order to calculate the 3D position coordinates from the images, corresponding points in the two images are required. When a point on the real space and its corresponding point on the image plane are located, the coordinates of the object can be calculated from the corresponding point through a geometric correlation between the image plane where the corresponding point is located and the space where the object is located. In this process, a camera matrix is required, and the relationship between the corresponding points and the camera matrix can be established. The camera matrix can be calculated from the basis matrix, but a sufficiently large number of correspondence points are required for accurate calculation. The process of calculating 3D position coordinates from the corresponding points of the image calculates a basis matrix from matching pairs of feature points, assumes a camera matrix from the calculated basis matrix, calculates the position coordinates of the GCP using the assumed camera matrix, , The homography matrix is calculated by comparing the measured coordinates with the calculated coordinates of the GCP, the assumed camera matrix is calibrated using the homography matrix, and the location coordinates of the corresponding point of the tracked object are calculated using the calibrated camera matrix.

A base matrix is used to calculate the camera matrix. A camera matrix is assumed from the basis matrix, and it is corrected using GCP. The first camera matrix is assumed from the relationship between the camera matrix and the base matrix, and the camera matrix is calibrated again using the geometric relationship between the position coordinates of the reconstructed scene and the GCP using the assumed camera matrix. If a unique elementary matrix is determined using several corresponding points in a stereo image, the 3D position of the camera and each corresponding point can be reconstructed using the elementary matrix and the corresponding corresponding points. In order to reconstruct the image in a form in which errors are maximally removed, correction of the camera matrix is required, and a homography matrix is used for this correction. The homography matrix can be calculated using actually measured 3D position coordinates and can be calculated using GCP.

If the corrected camera matrix and corresponding points of the image for calculating 3D position coordinates are given, 3D position coordinates can be calculated through triangulation. A linear triangulation method is used to calculate 3D position coordinates. The triangulation method is a method of calculating the positional coordinates of a point from a corresponding point created by projecting a point on an actual 3D space onto an image. Of course, each object can be tracked in various ways other than the above method, and each object can be simply tracked using YOLO or each object can be tracked using intelligent CCTV. If the object is identified and tracked in this way, based on the Industrial Accident Safety and Health Guidelines or government guidelines, it is defined which state is a high or low industrial accident risk state, and which actions are dangerous actions that violate the law. Hazard notifications can be provided immediately when such conditions or behaviors are seen.

The output unit 330 may transmit a risk event to at least one output terminal 400 when a risk event is output from at least one artificial intelligence algorithm. At least one output terminal 400 may output a risk event analyzed as an image of at least one input terminal 100 .

<Identification of collision risk>

A risk assessment index can be set for a reasonable risk assessment. The risk determination may be composed of an algorithm that determines the final risk using a distance-based index (Distance To Collision, DTC) and a time-based index (Time To Collision, TTC). Considering the TTC together, it is possible to define an algorithm that increases the risk if the speed is high even if the object is far away. In general, TTC should be given information such as the speed of heavy equipment. Even in the case of a forklift without a turning motion, it can be generally applied by replacing it with the moving speed of the equipment itself. TTC and DTC can be calculated by Equation 1 and Equation 2 below, where TTC is the collision prediction time with the object, x is the index using the distance information that can be stopped using the available deceleration and the distance to the current object indicates Since this is a general known technology, it will not be described in detail.

<Risk of equipment-worker collision>

The risk of collision between equipment and workers can use TTCP (Position) based on Euclidean distance by combining collectable location information based on the traditional TTC concept described above. The TTCP calculation formula and the calculation method of the indicators constituting the formula are as follows.

where lmin,t is the minimum Euclidean distance between the worker and the mobile device, (xp,t, yp,t) is the worker's position at any time t, (xe,t, ye,t) is any is the equipment position at time t.

t is the reference time, lr,t is the relative distance at time t, and vr,Δt is the reduction rate of relative distance at time t.

<Falling Object Accident>

In the work area where the quay crane is located in the apron, container loading and unloading work is performed by the spreader of the crane. This is an area where there is always a risk of accident due to reasons such as unpredictable equipment damage because repetitive work of transporting heavy containers by yard tractor or vice versa is performed. Accidents caused by these reasons can be divided into two main categories. First, as a countermeasure plan in terms of equipment, it is a plan to prevent accidents by essentially implementing in advance, such as regular and occasional inspection of equipment condition, life cycle management, and work environment management. Second, as a countermeasure from the perspective of workers, it is a plan to prevent accidents by correcting workers' behavior, such as preparatory exercise, prohibition of access to dangerous areas, prohibition of carelessness during dangerous work, and compliance with work procedures. Here, countermeasures on the equipment side can reduce the accident rate, but equipment obsolescence and poor inspection status are only at the level of a general checklist. Accident prevention is bound to be limited.

Therefore, the best way to prevent this type of accident is to induce workers to avoid the danger zone in advance from the point of view of workers. Falling object accident risk detection algorithm is an algorithm that informs the dangerous situation by detecting the approach to the dangerous area step by step. In order to inform the entry into the danger zone, the detection algorithm can be composed of a total of 5 steps. Each level can be composed of 5 levels: Level 0 (Blue, Safety), Level 1 (Green, Interest), Level 2 (Yellow, Caution), Level 3 (Orange, Warning), and Level 4 (Red, Danger). It can be set based on criteria that can be judged qualitatively and engineeringly.

The blind spot removal unit 340 may minimize blind spots in the field by analyzing an image received from at least one input terminal 100 . At least one input terminal 100 may include a wired and wireless CCTV, a smart glass, and an action cam. To this end, the blind spot removal unit 340 may use a link between a GIS technique using digital terrain models (DTM) and field information. Of course, Digital Elevation Models (DEMs), Digital Surface Models (DSMs), and Triangular Irregular Networks (TINs) may be used. The blind spot removal unit 340 inputs terrain data obtained from a topographical map or aerial photograph to a computer, converts the digital terrain model into a DTM model, and then automatically searches for visible and invisible areas by Viewshed Analysis Algorithm. The Visibility Analysis Basic Algorithm Model algorithm required for this is performed in two essential steps. The first is to retrieve the horizontal position of the grid DEM where the line of sight intersects the target from the observation point, and the second is to find the horizontal position of the DEM where the line of sight intersects the line of sight. to compare elevations.

If the latter is higher then it will continue looking for the next intersection, otherwise the target will be invisible from the observation point and will continue looking for another target. In this process, the method of inferring the elevation of the grid and the determination of the number and location of lines for line-of-sight comparison are very important factors. Similarly, the way the grid cell structure is interpreted with respect to observation points and targets and the overall DEM is always of importance. The second step in which the elevation is computed is the determining step, so it is an important step in finding visibility. In addition, by applying the height value through the building information linkage, it is possible to generate information on the visible area around the CCTV through analysis of the location and visibility information of the CCTV. Through this, the blind spot removal unit 340 can determine the non-visible area and route and display them on the map.

The display unit 350 outputs weather information, safety rule violations, and disaster risk calculated by a predetermined risk calculation formula. At least one output terminal 400 may include a monitor, a speaker, a laser indicator, and a user terminal that outputs a risk occurrence event. The laser indicator is a device that irradiates a light source with at least one color, and can be controlled to emit different colors for each risk level.

The prediction unit 360 detects whether or not there is a collision using a tracking algorithm that predicts and tracks the moving direction of workers, materials or equipment, and checks the displacement of the tilt by measuring the vertical and horizontal axes of the material or equipment to determine the possibility of overturning. Predictable.

The environmental detection unit 370 collects weather information to identify work hazards including temperature, precipitation, snowfall, wind speed, and fine dust in the case of outdoor work, and includes fall sections, constriction risk sections, and falling object risk sections. It is possible to check the presence or absence of safety facilities in the danger zone of the site.

The violation monitoring unit 380 may check violations in the workplace by applying at least one law or regulation, and compare and evaluate risks between workplaces by accumulating the number of occurrences of caution, alert, and risk in the same workplace as a database.

Hereinafter, an operation process according to the configuration of the above-described safety management service providing server of FIG. 2 will be described in detail with reference to FIGS. 3 and 4 as examples. However, it will be apparent that the embodiment is only any one of various embodiments of the present invention, and is not limited thereto.

Referring to FIG. 3, (a) the safety management service providing server 300 uses big data and artificial intelligence to prevent accidents by inserting real-time video as a query after modeling so that repetitive accidents for similar industrial accidents do not occur. An evacuation order can be given to workers (workers). In addition, (b) after receiving the input image, the safety management service providing server 300 extracts the object, calculates the moving direction and speed using the location, size, direction, speed, etc., and sends it to each output terminal 400. to provide. (c) The safety management service providing server 300 may monitor violations of laws and give warnings to workers, and (d) predict movement routes to prevent accidents in advance.

<Construction site>

It is assumed that the construction site as shown in FIG. 4a is divided into zones P1 to P4. In the P1 section, the movement of the ready-mixed concrete vehicle and the movement of workers are photographed at the input terminal 100 and transmitted to the safety management service providing server 300, and the safety management service providing server 300 calculates the vehicle movement speed and direction, , It predicts the movement direction of workers and determines the risk of accidents. In addition, it predicts the direction of vehicle movement and determines the risk of an accident. The output terminal 400 may display the operation risk level and the number of warnings generated in a section where vehicle, pedestrian, or roadway classification is insufficient or may be transmitted to a vehicle operator through an alarm system. In the event of a vehicle-to-vehicle collision, a hazard warning can be issued. You can also send text messages to administrators.

In the P2 section, it is a worker access prohibited section (machine operation section), and when it is confirmed that the worker is approaching, the input terminal 100 takes a picture and transmits it to the safety management service providing server 300, and the safety management service providing server 300 It checks whether or not the access restriction facility is installed (law violation), whether the worker is wearing protective gear (hard hat, safety shoes, seat belt), whether the worker is careless, and through the output terminal 400, the access restriction facility, the work risk level of the section, and a warning occurs The number of times is counted and notified by the monitor. In addition, an alarm is issued to the operator using an alarm system, a directional speaker, or a laser indicator, and a text message is sent to the manager's terminal.

In the P3 section, the input terminal 100 takes a picture of a worker working on the site floor and sends it to the safety management service providing server 300 . The safety management service providing server 300 checks whether workers wear protective gear, detects worker negligence, informs the occurrence of not wearing protective gear through a monitor at the output terminal 400, and updates the work risk level and the number of warnings. Alarm systems, directional speakers, and laser indicators alert workers and send text messages to managers. Assuming that the P4 section is in the process of raising a material by a hydro crane, the safety management service providing server 300 receives photographing data from the input terminal 100 in real time, and the safety management service providing server 300 determines whether the surroundings are controlled, that is, the material Check for violations of the law prohibiting entry into the lifting section, check whether the crane is leveled (overturn or collapse in case of poor leveling), submit a work plan for lifting surrounding materials, and check whether a signalman is placed (violation of the law). The safety management service providing server 300 outputs the horizontal destruction of the crane, the risk level of the corresponding section, and the number of warnings to the monitor through the output terminal 400, issues an alarm to all field workers through an alarm system, speaker, smart phone, etc. Send a text message to the manager and safety and health general manager on the smartphone.

<Manufacturing>

The safety management service providing server 300 receives the work of the worker around the rotating body as photographed data through the input terminal 100 in the P1 section, and the safety management service providing server 300 lacks safety measures to prevent the rotating body from rolling ( law violations), check not wearing safety gear (carelessness of workers), and check whether there are floor obstacles in the direction of workers (worker falls, danger). In addition, the safety management service providing server 300 provides the output terminal 400 with a monitor for the lack of safety measures to prevent rolling back of the rotating body, the work risk level in the work section, and the number of warnings, and alerts the worker with an alarm system, directional speaker, and laser. is issued, and a text message is sent to the administrator on the smartphone.

The safety management service providing server 300 checks that the worker is accessing the step occurrence section (fall risk section) through the input terminal 100 in the P2 section, and the safety management service providing server 300 checks whether or not the safety handrail is installed ( Violation of law), checks whether workers are wearing protective gear (carelessness of workers), provides the output terminal (400) for no railings, work risk in the corresponding section, and the number of warnings as a monitor, and provides workers with an alarm system, directional speaker, and laser indicator An alert is issued to the manager and a text message is sent to the manager via smartphone. In the P3 section, the safety management service providing server 300 receives and confirms the lifting operation of the ventry crane from the input terminal 100, and the safety management service providing server 300 controls whether or not the surroundings are controlled (violation of the law prohibiting entry to the lifting section of the material). , After submitting a work plan for lifting surrounding materials and confirming whether a signalman is placed (violation of the law), the output terminal 400 outputs whether or not the surroundings are controlled, the risk level of the section, and the number of warnings generated through the monitor, and sends a text message to the manager. there is.

In the P4 section, the safety management service providing server 300 confirms that deformation has occurred in the factory pillar (vertical member) through the input terminal 100, calculates the damage range in case of collapse, and predicts the risk of collapse by calculating the deformation . The safety management service providing server 300 provides vertical member deformation and destruction, work risk in the corresponding section, and the number of warnings through a monitor, which is an output terminal 400, and alerts workers in all workplaces with an alarm system, a directional speaker, and a smartphone. is issued, and a text message is sent to the manager and safety and health general manager via smartphone.

Matters not described in the smart intelligent safety management service provision method of FIGS. 2 to 4 are the same as or easily inferred from the description of the smart intelligent safety management service provision method through FIG. 1 above. Therefore, the following description is omitted.

5 is a diagram illustrating a process of transmitting and receiving data between each component included in the smart intelligent safety management service providing system of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process of transmitting and receiving data between each component will be described through FIG. 5, but the present application is not limited to such an embodiment, and according to various embodiments described above, It is obvious to those skilled in the art that a process of transmitting and receiving data may be changed.

Referring to FIG. 5 , the safety management service providing server receives an image received from at least one input terminal (S5100).

The safety management service providing server inputs the image as a query to at least one type of artificial intelligence algorithm (S5200), and when a dangerous event is output from the at least one artificial intelligence algorithm, the dangerous event is output to at least one output terminal. It transmits ㅎ Ha (S5300).

The order between the above-described steps (S5100 to S5300) is only an example, and is not limited thereto. That is, the order of the above-described steps (S5100 to S5300) may be mutually changed, and some of the steps may be simultaneously executed or deleted.

Matters not described in the smart intelligent safety management service provision method of FIG. 5 are the same as or easily inferred from the description of the smart intelligent safety management service provision method through FIGS. 1 to 4 above. Therefore, the following description is omitted.

The smart intelligent safety management service providing method according to an embodiment described with reference to FIG. 5 may be implemented in the form of a recording medium including instructions executable by a computer, such as an application or program module executed by a computer. . Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer readable media may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The above-described smart intelligent safety management service providing method according to an embodiment of the present invention may be executed by an application basically installed in the terminal (this may include a program included in a platform or operating system, etc. basically loaded in the terminal). and may be executed by an application (that is, a program) directly installed in the master terminal by a user through an application providing server such as an application store server, an application or a web server related to the corresponding service. In this sense, the smart intelligent safety management service providing method according to an embodiment of the present invention described above is implemented as an application (ie, a program) that is basically installed in a terminal or directly installed by a user and can be read by a computer such as a terminal. can be recorded on a recording medium.

The above description of the present invention is for illustrative purposes, and those skilled in the art can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts should be construed as being included in the scope of the present invention. do.

Claims (7)

In the smart intelligent safety management service providing system to prevent industrial accidents occurring at construction sites and manufacturing sites,
At least one input terminal for outputting in real time an image taken at the construction site and the manufacturing site using at least one type of camera;
At least one output terminal for outputting a risk event analyzed by the image taken at the construction site and the manufacturing site of the at least one input terminal; and
A receiving unit receiving the image taken at the construction site and manufacturing site received from the at least one input terminal; an input unit inputting the received image into at least one type of artificial intelligence algorithm as a query; When a dangerous event is output from at least one artificial intelligence algorithm, an output unit for transmitting the dangerous event to the at least one output terminal, and images taken at the construction site and manufacturing site received from the at least one input terminal By using a blind spot removal unit that analyzes and minimizes blind spots in the field, and a tracking algorithm that predicts and tracks the movement direction of workers, materials, and equipment at the construction site and manufacturing site, collision is detected, and the material or equipment is detected. A safety management service providing server including a predictor for predicting the possibility of overturning by checking the displacement of the inclination by measuring the vertical and horizontal axes of the;
A smart intelligent safety management service providing system that includes a.
delete According to claim 1,
The at least one output terminal includes a monitor, a speaker, a laser indicator, and a user terminal for outputting the dangerous event,
The safety management service providing server,
A display unit for outputting weather information, safety rule violations, and disaster risk calculated by a preset risk calculation formula;
A smart intelligent safety management service providing system further comprising a.
According to claim 3,
The laser indicator is a device for irradiating a light source with at least one color, and is controlled to be irradiated with different colors for each risk level.
delete According to claim 1,
The safety management service providing server,
Weather information is collected to identify work hazards including temperature, precipitation, snowfall, wind speed, and fine dust in the case of outdoor work, and safety facilities are installed in risk zones within the site, including fall zones, constriction risk zones, and falling objects risk zones. Environment sensing unit to check the presence or absence;
A smart intelligent safety management service providing system further comprising a.
According to claim 1,
The safety management service providing server,
A violation monitoring unit that checks for violations in the workplace by applying at least one law and regulation, and compares and evaluates the risk between workplaces by accumulating the number of occurrences of caution, alertness, and risk in the same workplace as a database;
A smart intelligent safety management service providing system further comprising a.

Images