KR102511093B1 - Smart city elevator crime prevention system - Google Patents

Abstract

An elevator crime prevention system according to an embodiment includes at least one sensor, a platform server connected to the at least one sensor for data communication, and a control unit operatively connected to the at least one sensor, The controller obtains motion information and/or sound information generated inside the elevator by controlling the at least one sensor, and based on the motion information and/or the sound information, the motion information and/or the sound. It is determined whether the information satisfies a specified condition, and when the motion information and/or the sound information satisfy the specified condition, alert data may be transmitted to the client terminal through the platform server.

Description

Smart City Elevator Crime Prevention System {SMART CITY ELEVATOR CRIME PREVENTION SYSTEM}

Various embodiments disclosed in this document relate to a smart city elevator crime prevention system. More specifically, it relates to a system for preventing crime by recognizing motion and/or voice in an elevator and monitoring crime in real time.

In general, a surveillance system that monitors a specific place through surveillance means and enables countermeasures or post-confirmation when an abnormality is found is introduced in various places such as entrances, parking lots, buildings, industrial sites, and residential areas where security is important. come. Through this, as security is improved, access control becomes easier, and the effect of reducing the crime rate is proven, it is gradually becoming more diverse and generalized.

In general, these monitoring means are generally made in a way to acquire image and video information, such as CCTV, and to identify objects and actions based on the acquired information.

In particular, deep learning and machine learning algorithms are currently being studied as technologies that accept, analyze, and interpret a lot of information through computers, and as these technologies are applied to surveillance systems, by learning objects, actions, and voices, crimes occur The technology to determine whether or not is being applied.

However, there are cases in which conventional surveillance systems stop at recording video information and are limited to black box-type video analysis systems for post-causal analysis rather than preventive purposes due to lack of manpower for real-time monitoring.

In addition, since existing surveillance systems acquire only video information and do not acquire sound information, there may be limitations in monitoring crimes such as intimidation occurring in the blind spot of video information acquisition devices such as CCTVs through voice.

In addition, the existing surveillance system must be detected by manpower, and if the surveillance manpower is inexperienced in coping with the occurrence of a crime, the surveillance system alone may be insufficient for crime prevention.

The technical tasks to be achieved in the present disclosure are not limited to the technical tasks mentioned above, and other technical tasks not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

An elevator crime prevention system according to an embodiment includes at least one sensor, a platform server connected to the at least one sensor for data communication, and a control unit operatively connected to the at least one sensor, The controller obtains motion information and/or sound information generated inside the elevator by controlling the at least one sensor, and based on the motion information and/or the sound information, the motion information and/or the sound. It is determined whether the information satisfies a specified condition, and when the operation information and/or the sound information satisfy the specified condition, alert data may be transmitted to the client terminal through the platform server.

An elevator crime prevention method according to an embodiment is an operation of obtaining operation information and / or sound information occurring inside the elevator using at least one sensor, based on the operation information and / or the sound information, In response to determining whether the operation information and/or the sound information satisfies the specified condition, and determining that the operation information and/or the sound information satisfy the specified condition, the client terminal via the platform server. It may include an operation of transmitting alert data to .
In addition, in the elevator crime prevention system, at least one sensor; a platform server connected to the at least one sensor to enable data communication; and a control unit operatively connected to the at least one sensor, wherein the control unit obtains motion information and sound information generated inside the elevator by controlling the at least one sensor, and the motion information In order to obtain, the at least one sensor includes at least one of a camera or a CCTV or a motion detection sensor, and to obtain the sound information, the at least one sensor uses at least one of a sound detection sensor or a voice recognition sensor. and determining, based on the motion information and the sound information, whether the motion information and the sound information satisfy a specified condition, wherein the specified condition determines whether or not a crime has occurred based on the motion information and the sound information. and if the client terminal can be referred to at least one of PC, Web, and Application of a crime-related agency, and the operation information and the sound information satisfy the specified conditions, an alarm is sent to the client terminal through the platform server. transmits data, and further includes an artificial intelligence learning module, wherein the control unit learns the motion information and the sound information obtained by controlling the at least one sensor through the artificial intelligence learning module, and the learning result Based on, monitoring whether a crime occurs in the elevator, and estimating the posture of a person inside the elevator by applying DeepPose and HRNET2 algorithms to the data detected through the at least one sensor, At least some of possible fights, falls, and abandonments are estimated, postures are estimated through joint coordinates, 3D postures are estimated through depth information, and R(2+1) is applied to the data detected through the at least one sensor. D and I3D algorithms are applied to estimate the behavior of the person inside the elevator, instantaneous motion and behavior are estimated, and Mask R is applied to the data detected through the at least one sensor. -Apply the CNN algorithm to estimate the hand motion of the person inside the elevator, detect and estimate the hand motion by projecting three-dimensional coordinates on 16 finger joints, and apply the Kospeech algorithm to the data detected through the at least one sensor Applied to estimate the sound generated inside the elevator, detect and estimate the voice or sound promised in advance through acoustic model and language model learning, and for the motion information, the DeepPose and HRNET2 algorithms, the R ( 2+1) D and I3D algorithms, and the Mask R-CNN algorithm are used to estimate the human posture, human action, and human hand motion, and the sound is estimated using the Kospeech algorithm for the sound information and a network video recorder (NVR) server communicatively connected to the at least one sensor to receive and store the operation information and the sound information, wherein the platform server includes the operation information and the sound from the NVR server. Receive data including information, wherein the specified condition includes whether the motion information or the sound information matches a predetermined motion or a predetermined sound, wherein the predetermined motion is at least one of a posture, an action, and a hand motion. and a speaker disposed inside the elevator, wherein the control unit outputs an alarm sound to the inside of the elevator through the speaker when the operation information and the sound information satisfy specified conditions. to be characterized
In addition, in the elevator crime prevention method, using at least one sensor, obtaining operation information and sound information generated inside the elevator; determining whether the motion information and the sound information satisfy specified conditions based on the motion information and the sound information; To obtain the motion information, the at least one sensor includes at least one of a camera or a CCTV or a motion sensor, and to obtain the sound information, the at least one sensor is a sound sensor or a voice recognition sensor. and transmitting alert data to a client terminal through a platform server in response to determining that the operation information and the sound information satisfy the specified condition, and the operation information and the sound include at least one. The operation of determining whether information satisfies a specified condition further includes an artificial intelligence learning module, wherein the specified condition determines whether or not a crime has occurred based on the motion information and the sound information, and the client terminal is related to a crime. What can be referred to as at least one of the institution's PC, Web, and Application, wherein the control unit learns the motion information and the sound information obtained by controlling the at least one sensor through the artificial intelligence learning module, and Based on the learning result, monitoring whether a crime occurs in the elevator, and estimating the posture of a person inside the elevator by applying DeepPose and HRNET2 algorithms to the data detected through the at least one sensor, the elevator Estimating at least some of possible fights, falls, and abandonments, estimating posture through coordinates of joints, estimating 3D posture through depth information, and adding R(2+ to data detected through the at least one sensor) 1) Estimating the behavior of a person inside the elevator by applying D and I3D algorithms, estimating instantaneous motion and behavior, and applying the Mask R-CNN algorithm to data sensed through the at least one sensor to estimate the elevator The hand motion of the person inside is estimated, but the hand motion is detected and estimated by projecting three-dimensional coordinates on 16 finger joints, and Kospeech algorithm is applied to the data detected through the at least one sensor. A gorism is applied to estimate the sound generated inside the elevator, but through acoustic model and language model learning, pre-promised voice or sound is detected and estimated, and the DeepPose and HRNET2 algorithms and the The human posture, human action, and human hand motion are estimated using the R(2+1)D and I3D algorithms and the Mask R-CNN algorithm, respectively, and the Kospeech algorithm is used to estimate the sound information. The operation of estimating and determining whether the motion information and the sound information satisfy a specified condition includes an operation of determining whether the motion information matches at least one of a predetermined posture, action, and hand motion, The operation of determining whether the operation information and the sound information satisfy a specified condition includes determining whether the sound information matches a pre-specified voice, and the operation information and the sound information satisfy the specified condition. When satisfied, it characterized in that it comprises an operation of outputting an alarm sound to the inside of the elevator through a speaker.

According to various embodiments of the present disclosure, the occurrence of a crime can be immediately detected by collecting and monitoring crime information in an elevator in real time.

In addition, according to various embodiments of the present disclosure, it is possible to quickly respond to crime by detecting the occurrence of a crime in an elevator in real time and transmitting it to a related institution.

In addition, according to various embodiments of the present disclosure, it is possible to prevent crime and additional crime by detecting crime in the elevator in real time and outputting a warning message in response thereto.

The effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. will be.

Also other aspects as described above, features and benefits of certain preferred embodiments of the present invention will become more apparent from the following description taken in conjunction with the accompanying drawings.
1 shows an elevator crime prevention system according to an embodiment.
2 illustrates a structure for transmitting detected information to a client terminal according to an embodiment.
3A illustrates an algorithm for detecting and estimating a posture according to an embodiment.
3B depicts an algorithm for identifying actions, according to one embodiment.
3C illustrates an algorithm for detecting and estimating a hand gesture according to an embodiment.
4 illustrates an algorithm for recognizing voice information according to an embodiment.
5 is a flowchart illustrating an elevator crime prevention method according to an embodiment.
6 is a flowchart illustrating an operation of outputting an alarm sound through a speaker according to an exemplary embodiment.
In connection with the description of the drawings, like reference numerals may be used for identical or substantially identical elements.

Advantages and features of the present invention, and methods for achieving them, will become clear with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention belongs It is provided to fully inform the holder of the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals designate like elements throughout the specification.

In describing the embodiments of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In addition, the terms to be described below are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to the intention or custom of a user or operator. Therefore, the definition should be made based on the contents throughout this specification.

At this time, it will be understood that each block of the process flow chart diagrams and combinations of the flow chart diagrams can be performed by computer program instructions. These computer program instructions may be embodied in a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, so that the instructions executed by the processor of the computer or other programmable data processing equipment are described in the flowchart block(s). It creates means to perform functions. These computer program instructions may also be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular way, such that the computer usable or computer readable memory The instructions stored in are also capable of producing an article of manufacture containing instruction means that perform the functions described in the flowchart block(s). The computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operational steps are performed on the computer or other programmable data processing equipment to create a computer-executed process to generate computer or other programmable data processing equipment. Instructions for performing processing equipment may also provide steps for performing the functions described in the flowchart block(s).

Additionally, each block may represent a module, segment, or portion of code that includes one or more executable instructions for executing specified logical function(s). It should also be noted that in some alternative implementations it is possible for the functions mentioned in the blocks to occur out of order. For example, two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in reverse order depending on their function.

At this time, the term '~unit' used in this embodiment means software or a hardware component such as a field-programmable gate array (FPGA) or application specific integrated circuit (ASIC), and what role does '~unit' have? perform them However, '~ part' is not limited to software or hardware. '~bu' may be configured to be in an addressable storage medium and may be configured to reproduce one or more processors. Therefore, 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 describing the embodiments of the present invention in detail, an example of a specific system will be the main target, but the main subject matter to be claimed in this specification extends the scope disclosed herein to other communication systems and services having a similar technical background. It can be applied within a range that does not deviate greatly, and this will be possible with the judgment of those skilled in the art.

1 shows an elevator crime prevention system according to an embodiment.

Referring to FIG. 1 , an elevator crime prevention system 100 (hereinafter, crime prevention system) according to an embodiment includes at least one sensor 110 and a network video recording (NVR) connected to communicate with the at least one sensor. It may include a server 120, a platform server 130, and a client terminal 140 connected to the platform server 130 to enable data communication.

According to an embodiment, the crime prevention system 100 may obtain motion information and/or sound information through at least one sensor 110 . According to one embodiment, the crime prevention system 100 may detect motion and/or sound inside the elevator using at least one sensor 110 .

According to one embodiment, at least one sensor 110 may be referred to as a camera or closed circuit television (CCTV). According to another embodiment, at least one sensor 110 may be referred to as one of a motion detection sensor, a sound detection sensor, or a voice recognition sensor, but is not limited thereto. For example, the crime prevention system 100 may detect motion inside an elevator through a motion sensor. For another example, the crime prevention system 100 may detect at least one of a posture, hand motion, or action of a person present inside the elevator through CCTV. As another example, the crime prevention system 100 may detect sound generated inside the elevator through a sound sensor.

According to an embodiment, the crime prevention system 100 may transmit motion information and/or sound information acquired through at least one sensor 110 to the NVR server 120 . According to an embodiment, the NVR server 120 may compress and store acquired motion information and/or sound information as digital data. According to one embodiment, the NVR server 120 may edit and output motion information and/or sound information stored in digital form. According to one embodiment, the NVR server 120 may transmit motion information and/or sound information stored in digital form to the platform server 130 .

According to one embodiment, the crime prevention system 100 may include a platform server 130 connected to the NVR server 120 to enable data communication. The platform server 130 according to an embodiment may receive motion information and/or sound information converted into digital form from the NVR server 120 . For example, the NVR server 120 may transmit data to the platform server 130 using real time streaming protocol (RTSP) communication.

According to an embodiment, the platform server 130 may determine whether a crime has occurred inside the elevator based on motion information and/or sound information received from the NVR server 120 . A detailed description of this will be described later with reference to FIG. 5 .

According to an embodiment, the platform server 130 may generate situation data and alarm data based on motion information and/or sound information received from the NVR server 120 . According to an embodiment, the platform server 130 may generate at least some of situation data and alarm data by analyzing motion information and/or sound information received from the NVR server 120 . For example, the platform server 130 may include an artificial intelligence learning module (eg, a machine learning model) and learn motion information and/or sound information received from the NVR server 120 through the artificial intelligence learning module. there is. For example, the platform server 130 learns motion information and/or sound information received from the NVR server 120 using an artificial intelligence learning module (eg, a machine learning model), and based on the learning result, the elevator It can determine whether a crime has occurred inside. For example, the platform server 130 learns motion information and/or sound information received from the NVR server 120 using an artificial intelligence learning module (eg, a machine learning model), and based on the learning result, a situation At least some of data and alert data may be generated.

2 illustrates a structure for transmitting detected information to a client terminal according to an embodiment.

Referring to FIGS. 1 and 2 together, the platform server 130 according to an embodiment may be referred to as either a local server or a cloud server. According to one embodiment, the platform server 130 may store, edit, and transmit data received from at least one sensor 110 or NVR server 120 . For example, the platform server 130 may transmit data in a forward proxy method of receiving a request from an internal network and transmitting data to the client terminal 140 by connecting to the Internet.

According to one embodiment, the platform server 130 may transmit at least some of the situation data and alert data to the client terminal 140 . According to an embodiment, the platform server 130 may transmit at least some of the situation data and alarm data generated according to the analysis result to the client terminal 140 through an application built in the web server. According to an embodiment, the crime prevention system 100 may transmit at least some of situation data and alarm data to the client terminal 140 through the platform server 130 when recognizing that a crime occurs inside the elevator. .

According to one embodiment, the client terminal 140 may refer to at least one of a PC (personal computer), Web, and APP (application) of a crime-related institution (eg, a police station or a fire station), but is not limited thereto, and various types of It may be referred to as an electronic device of.

According to an embodiment, the crime prevention system 100 obtains motion information and/or sound information through at least one sensor 110, determines whether a crime has occurred based on the obtained information, and determines that a crime has occurred. If determined, the situation data and alert data may be transmitted to the client terminal 140 .

3A illustrates an algorithm for detecting and estimating a posture according to an embodiment. 3B depicts an algorithm for identifying actions, according to one embodiment. 3C illustrates an algorithm for detecting and estimating a hand gesture according to an embodiment.

Referring to FIGS. 3A to 3C , the crime prevention system 100 according to an embodiment may detect an operation of an event occurring inside an elevator through at least one sensor 110 . According to an embodiment, the crime prevention system 100 may detect at least some of the posture, behavior, and hand motion of a person present in the elevator through at least one sensor 110 .

Referring to FIG. 3A , the crime prevention system 100 according to an embodiment may obtain information about a person's posture inside an elevator using at least one sensor 110 (eg, a motion detection sensor).

According to an embodiment, the crime prevention system 100 may estimate a posture based on data sensed through at least one sensor 110 . For example, the crime prevention system 100 may estimate the posture of a person inside the elevator by applying DeepPose and HRNET2 algorithms of the mmPose package to data sensed through at least one sensor 110 . For example, the crime prevention system 100 estimates at least some of fights, falls, and abandonments that may occur in an elevator by applying DeepPose and HRNET2 algorithms of the mmPose package to data detected through at least one sensor 110. can For example, the crime prevention system 100 applies the DeepPose and HRNET2 algorithms of the mmPose package to data detected through at least one sensor 110, thereby estimating the posture through joint coordinates and 3D data through depth information. position can be estimated. According to an embodiment, the crime prevention system 100 may detect a crime occurrence situation by estimating a posture by applying DeepPose and HRNET2 algorithms of the mmPose package to data sensed through at least one sensor 110 .

Referring to FIG. 3B , the crime prevention system 100 according to an embodiment may obtain information about a person's behavior in an elevator using at least one sensor 110 (eg, a motion detection sensor).

According to an embodiment, the crime prevention system 100 may estimate a person's behavior inside the elevator based on data sensed through at least one sensor 110 . For example, the crime prevention system 100 applies the R(2+1)D and I3D algorithms of the mmAction2 package to data sensed through at least one sensor 110 to estimate the behavior of a person inside the elevator. can For example, the crime prevention system 100 complementarily uses DeepPose and HRNET2 algorithms of the mmPose package and R(2+1)D and I3D algorithms of the mmAction2 package on data detected through at least one sensor 110. By applying it, instantaneous motions and behaviors can be estimated with high reliability.

Referring to FIG. 3C , the crime prevention system 100 according to an embodiment may obtain information about a person's hand motion inside an elevator using at least one sensor 110 (eg, a motion detection sensor).

According to an embodiment, the crime prevention system 100 may estimate a person's hand motion inside the elevator based on data sensed through at least one sensor 110 . For example, the crime prevention system 100 may estimate the hand motion of a person inside the elevator by applying the Mask R-CNN algorithm of the Mmdetection package to data sensed through at least one sensor 110 . For example, the crime prevention system detects and estimates hand motions by projecting three-dimensional coordinates for 16 finger joints by applying the Mask R-CNN algorithm of the Mmdetection package to data detected through at least one sensor 110. can do. According to an embodiment, the crime prevention system 100 applies the Mask R-CNN algorithm of the Mmdetection package to the data sensed through at least one sensor 110 to detect and estimate a specific hand motion promised in advance, thereby estimating a crime. occurrence can be detected.

4 illustrates an algorithm for recognizing voice information according to an embodiment.

Referring to FIG. 4 , the crime prevention system 100 according to an embodiment obtains information about sound generated inside an elevator using at least one sensor 110 (eg, a sound detection sensor, a voice recognition sensor). can do.

According to an embodiment, the crime prevention system 100 may estimate the sound generated inside the elevator based on data detected through at least one sensor 110 . For example, the crime prevention system 100 may estimate the sound generated inside the elevator by applying the Kospeech algorithm of the Joint-ctc-attention package to data sensed through at least one sensor 110 . For example, the crime prevention system applies the Kospeech algorithm of the Joint-ctc-attention package to data sensed through at least one sensor 110, and through learning an acoustic model and a language model, a previously promised voice or sound is obtained. can be detected and estimated. According to an embodiment, the crime prevention system 100 detects a voice or a specific sound generated in an elevator by applying the Kospeech algorithm of the Joint-ctc-attention package to data detected through at least one sensor 110. , it is possible to detect the crime occurrence situation by estimating.

5 is a flowchart illustrating an elevator crime prevention method according to an embodiment.

Referring to FIG. 5 , in the crime prevention system 100 according to an embodiment, when a crime is detected according to motion and/or sound generated inside the elevator through at least one sensor 110, alarm data and/or Condition data may be transmitted to the client terminal 140 .

According to an embodiment, the crime prevention system 100 (or control unit) may obtain motion information and/or sound information generated inside the elevator using at least one sensor 110 in operation 501. According to an embodiment, the crime prevention system 100 may obtain motion information or sound information based on motion or sound inside the elevator sensed using at least one sensor 110 in operation 501 . For example, the crime prevention system 100 may detect a voice generated inside an elevator using a voice recognition sensor. For another example, the crime prevention system 100 may detect at least some of a person's motion, posture, or hand motion inside the elevator using a motion sensor.

According to an embodiment, in operation 503, the crime prevention system 100 (or the control unit) may determine whether motion information and/or sound information obtained through operation 501 satisfy a specified condition. According to an embodiment, the specified condition may be referred to as a pre-specified crime occurrence situation. For example, in operation 503, the crime prevention system 100 may determine whether motion information or sound information matches a predetermined motion or a predetermined sound. For another example, in operation 503, the crime prevention system 100 may determine whether motion information or sound information matches a crime occurrence situation derived through learning of an artificial intelligence learning model.

According to an embodiment, when the crime prevention system 100 (or control unit) determines that motion information or sound information satisfies a pre-specified condition through operation 503, through the platform server 130 in operation 505, the client terminal ( 140) may transmit alarm data. According to an embodiment, the crime prevention system 100 transmits at least one of alert data and situation data generated through the platform server 130 in operation 505 when a crime is detected through operation 503 to the client terminal 140. can be sent to An operation in which the crime prevention system 100 according to an embodiment transmits alert data and/or situation data to the client terminal 140 in operation 505 may be referred to through configurations of FIGS. 1 and 2 .

6 is a flowchart illustrating an operation of outputting an alarm sound through a speaker according to an exemplary embodiment.

5 and 6, the crime prevention system 100 according to an embodiment transmits alarm data to the client terminal 140 when a crime is detected based on an operation or sound inside the elevator, and the elevator An alarm sound can be output to

According to one embodiment, the crime prevention system 100 may further include at least one speaker disposed inside the elevator. According to one embodiment, the crime prevention system 100 may output an alarm sound through at least one speaker disposed inside the elevator in operation 601 . According to one embodiment, the crime prevention system 100 may output an alarm sound to the inside of the elevator through at least one speaker disposed inside the elevator in operation 601 .

According to one embodiment, the crime prevention system 100 outputs an alarm sound to the inside of the elevator through at least one speaker disposed inside the elevator in operation 601, thereby inducing a crime to be stopped or preventing additional crime.

Elevator crime prevention system 100 according to an embodiment is operatively connected to at least one sensor, a platform server 130 connected to enable data communication with the at least one sensor, and the at least one sensor And a control unit that obtains operation information and / or sound information generated inside the elevator by controlling the at least one sensor, and based on the operation information and / or the sound information, the operation information It is determined whether the information and/or the sound information satisfies a specified condition, and when the operation information and/or the sound information satisfies the specified condition, the client terminal 140 is transmitted through the platform server 130. Alert data can be transmitted.

The elevator crime prevention system 100 according to an embodiment includes a network video recorder (NVR) server 120 that is communicatively connected to the at least one sensor and receives and stores the operation information and/or the sound information. and the platform server 130 may receive data including the operation information and/or the sound information from the NVR server 120 .

According to an embodiment, the designated condition may include whether the motion information or the sound information matches a predetermined motion or a predetermined sound, respectively.

According to an embodiment, the predetermined motion may include at least one of a posture, a behavior, and a hand motion.

According to an embodiment, the crime prevention system 100 further includes an artificial intelligence learning module, and the control unit controls the at least one sensor through the artificial intelligence learning module, and/or the operation information and/or the information obtained by controlling the at least one sensor. It is possible to learn sound information and monitor whether or not a crime occurs in the elevator based on the learning result.

According to one embodiment, the crime prevention system 100 further includes a speaker disposed inside the elevator, and the controller, when the operation information and/or the sound information satisfies a specified condition, through the speaker An alarm sound can be output to the inside of the elevator.

An elevator crime prevention method according to an embodiment is an operation of obtaining operation information and / or sound information occurring inside the elevator using at least one sensor, based on the operation information and / or the sound information, In response to determining whether the operation information and/or the sound information satisfies the specified condition, and determining that the motion information and/or the sound information satisfy the specified condition, the platform server 130 It may include an operation of transmitting alert data to the client terminal 140 through.

According to an embodiment, the operation of determining whether the motion information and/or the sound information satisfies a specified condition is an operation of determining whether the motion information matches at least one of a pre-specified posture, behavior, and hand motion. can include

According to an embodiment, the operation of determining whether the operation information and/or the sound information satisfies a specified condition may include an operation of determining whether the sound information matches a pre-specified voice.

According to an embodiment, when the operation information and/or the sound information satisfies a specified condition, an operation of outputting an alarm sound to the inside of the elevator through a speaker may be included.

As such, various embodiments of the present invention may be implemented as computer readable codes in a computer readable recording medium from a specific point of view. A computer readable recording medium is any data storage device capable of storing data readable by a computer system. Examples of computer readable recording media include read only memory (ROM), random access memory (RAM), and compact disk-read only memory (CD-ROM). ), magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data transmission over the Internet). The computer readable recording medium may also be distributed across networked computer systems, so that computer readable code is stored and executed in a distributed manner. In addition, functional programs, code, and code segments for achieving various embodiments of the present invention can be easily interpreted by programmers skilled in the field to which the present invention is applied.

It will also be appreciated that the apparatus and method according to various embodiments of the present invention can be realized in the form of hardware, software, or a combination of hardware and software. Such software may include, for example, volatile or non-volatile storage devices such as ROM, whether removable or rewritable, or memory, such as RAM, memory chips, devices or integrated circuits, or For example, it may be stored in an optically or magnetically recordable and machine (eg, computer) readable storage medium such as a compact disk (CD), DVD, magnetic disk, or magnetic tape. A method according to various embodiments of the present invention may be implemented by a computer or portable terminal including a control module and a memory, and the memory stores a program or programs including instructions for implementing the embodiments of the present invention. It will be appreciated that this is an example of a machine-readable storage medium suitable for

Accordingly, the present invention includes a program including code for implementing the device or method described in the claims of this specification and a storage medium readable by a machine (such as a computer) storing such a program. In addition, such a program may be transmitted electronically through any medium, such as a communication signal transmitted through a wired or wireless connection, and the present invention appropriately includes equivalents thereto.

Embodiments of the present invention disclosed in the present specification and drawings are merely presented as specific examples to easily explain the technical content of the present invention and help understanding of the present invention, and are not intended to limit the scope of the present invention. In addition, the embodiments according to the present invention described above are merely exemplary, and those skilled in the art will understand that various modifications and embodiments of equivalent range are possible therefrom. Therefore, the true technical protection scope of the present invention should be defined by the following claims.

Claims (10)

In the elevator crime prevention system,
at least one sensor;
a platform server connected to the at least one sensor to enable data communication; and
A control unit operatively connected with the at least one sensor,
The control unit:
Obtaining motion information and sound information generated inside the elevator by controlling the at least one sensor;
To obtain the motion information, the at least one sensor includes at least one of a camera or a CCTV or a motion sensor,
To obtain the sound information, the at least one sensor includes at least one of a sound detection sensor and a voice recognition sensor,
Based on the motion information and the sound information, determining whether the motion information and the sound information satisfy a specified condition;
The specified condition determines whether a crime has occurred based on the motion information and the sound information,
The client terminal may refer to at least one of a PC, Web, and Application of a crime-related institution;
When the operation information and the sound information satisfy the specified condition, alert data is transmitted to a client terminal through the platform server;
It further includes an artificial intelligence learning module;
The control unit learns, through the artificial intelligence learning module, the motion information and the sound information obtained by controlling the at least one sensor, and based on the learning result, determines whether a crime occurs in the elevator. monitoring,
Estimating the posture of a person inside the elevator by applying DeepPose and HRNET2 algorithms to the data detected through the at least one sensor, estimating at least some of fights, falls, and abandonments that may occur in the elevator, and the coordinates of the joints estimating posture through depth information, estimating 3D posture through depth information,
Estimating the behavior of a person inside the elevator by applying R(2+1)D and I3D algorithms to the data sensed through the at least one sensor, estimating instantaneous motion and behavior,
The mask R-CNN algorithm is applied to the data detected through the at least one sensor to estimate the hand motion of the person inside the elevator, and the hand motion is detected and estimated by projecting three-dimensional coordinates on 16 finger joints,
Estimate the sound generated inside the elevator by applying the Kospeech algorithm to the data detected through the at least one sensor, and detect and estimate a previously promised voice or sound through learning an acoustic model and a language model,
With respect to the motion information, the DeepPose and HRNET2 algorithms, the R (2 + 1) D and I3D algorithms, and the Mask R-CNN algorithm are used to estimate the human posture, human behavior, and human hand gestures, , The sound is estimated using the Kospeech algorithm for the sound information,
A network video recorder (NVR) server connected to the at least one sensor to be communicatively connected to receive and store the operation information and the sound information;
The platform server receives data including the operation information and the sound information from the NVR server,
The specified condition includes whether the motion information or the sound information matches a pre-specified motion or a pre-specified sound, respectively;
The predetermined motion includes at least one of a posture, an action, and a hand motion,
Further comprising a speaker disposed inside the elevator,
The control unit outputs an alarm sound to the inside of the elevator through the speaker when the operation information and the sound information satisfy specified conditions.
delete delete delete delete delete In the elevator crime prevention method,
obtaining operation information and sound information generated inside the elevator by using at least one sensor;
determining whether the motion information and the sound information satisfy specified conditions based on the motion information and the sound information;
To obtain the motion information, the at least one sensor includes at least one of a camera or a CCTV or a motion sensor,
To obtain the sound information, the at least one sensor includes at least one of a sound detection sensor and a voice recognition sensor,
Transmitting alert data to a client terminal through a platform server in response to determining that the operation information and the sound information satisfy the specified condition;
The operation of determining whether the motion information and the sound information satisfy a specified condition further includes an artificial intelligence learning module,
The specified condition determines whether a crime has occurred based on the motion information and the sound information,
The client terminal may refer to at least one of a PC, Web, and Application of a crime-related institution;
The control unit learns, through the artificial intelligence learning module, the motion information and the sound information obtained by controlling the at least one sensor, and based on the learning result, determines whether a crime occurs in the elevator. monitoring,
Estimating the posture of a person inside the elevator by applying DeepPose and HRNET2 algorithms to the data detected through the at least one sensor, estimating at least some of fights, falls, and abandonments that may occur in the elevator, and the coordinates of the joints estimating posture through depth information, estimating 3D posture through depth information,
Estimating the behavior of a person inside the elevator by applying R(2+1)D and I3D algorithms to the data sensed through the at least one sensor, estimating instantaneous motion and behavior,
The mask R-CNN algorithm is applied to the data detected through the at least one sensor to estimate the hand motion of the person inside the elevator, and the hand motion is detected and estimated by projecting three-dimensional coordinates on 16 finger joints,
Estimate the sound generated inside the elevator by applying the Kospeech algorithm to the data detected through the at least one sensor, and detect and estimate a previously promised voice or sound through learning an acoustic model and a language model,
With respect to the motion information, the DeepPose and HRNET2 algorithms, the R (2 + 1) D and I3D algorithms, and the Mask R-CNN algorithm are used to estimate the human posture, human behavior, and human hand gestures, , The sound is estimated using the Kospeech algorithm for the sound information,
The operation of determining whether the motion information and the sound information satisfy a specified condition includes determining whether the motion information matches at least one of a predetermined posture, action, and hand motion;
The operation of determining whether the operation information and the sound information satisfy a specified condition includes determining whether the sound information matches a pre-specified voice,
Elevator crime prevention method comprising outputting an alarm sound to the inside of the elevator through a speaker when the operation information and the sound information satisfy a specified condition.
delete delete delete

Images