KR20220000493A - Method for providing modularized artificial intelligence model platform service - Google Patents

Abstract

A modularized artificial intelligence model providing service providing method, a device thereof and a computer program thereof are provided. According to various embodiments of the present invention, the modularized artificial intelligence model providing service providing method is performed by a computing device. The method comprises the following steps of: obtaining a plurality of artificial intelligence modules generated by modularizing each of a plurality of artificial intelligence models; and providing one or more artificial intelligence modules corresponding to an artificial intelligence module providing request among the plurality of artificial intelligence modules in response to obtaining the artificial intelligence module providing request from a user.

Description

How to provide AI module platform service {METHOD FOR PROVIDING MODULARIZED ARTIFICIAL INTELLIGENCE MODEL PLATFORM SERVICE}

Various embodiments of the present invention relate to a modularized artificial intelligence model platform service providing method, apparatus, and computer program.

Artificial intelligence refers to a technology that realizes human learning ability, reasoning ability, perceptual ability, and natural language understanding ability through computer programs.

In addition, artificial intelligence is a field of computer engineering and information technology that studies how computers can do the thinking, learning, and self-development that can be done by human intelligence. say to make it possible

The technology for recognizing objects from images is being actively studied in various fields such as image processing, pattern recognition, computer vision, neural networks, and artificial intelligence, and has numerous commercial and legal applications.

In general, in order to use an artificial intelligence model to which artificial intelligence is applied, a manpower to develop an artificial intelligence model is deployed, an artificial intelligence model is produced from a company that develops an artificial intelligence model, or a desired artificial intelligence model is selected from a platform that provides a plurality of artificial intelligence models. The way it is provided is taken.

However, in the case of a method of receiving a desired artificial intelligence model from a platform that provides a plurality of artificial intelligence models, there is a problem in that it is difficult to provide an artificial intelligence model that fits the user's needs.

In addition, if manpower to develop an AI model is deployed or an AI model is produced by a company that develops an AI model, an AI model that fits the user's needs can be provided, but there is a problem that additional costs are incurred for this. have.

Korean Patent Registration No. 10-2064282 (2020.01.03)

The problem to be solved by the present invention is to provide an artificial intelligence module generated by modularizing an artificial intelligence model, but by providing the authority for the artificial intelligence module together with the artificial intelligence module, the user can directly modify and combine the artificial intelligence module. It is to provide a method, apparatus and computer program for providing a modularized artificial intelligence model providing service that enables the use of an artificial intelligence module suitable for the purpose of use.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

Modularized artificial intelligence model providing service providing method according to an embodiment of the present invention for solving the above problems, in a method performed by a computing device, generated by modularity of each of a plurality of artificial intelligence models (Modularity) may include obtaining a plurality of artificial intelligence modules and providing one or more artificial intelligence modules corresponding to the artificial intelligence module provision request among the plurality of artificial intelligence modules in response to obtaining an artificial intelligence module provision request from the user. have.

In various embodiments, providing the one or more artificial intelligence modules may include providing authority to the one or more artificial intelligence modules together with the one or more artificial intelligence modules.

In various embodiments, the obtaining of the plurality of artificial intelligence modules includes: a second artificial intelligence module generated by modifying the first artificial intelligence module from the user in response to the first artificial intelligence module provided to the user; obtaining a third artificial intelligence module generated by combining at least a part of the first artificial intelligence module and the second artificial intelligence module, and storing the second artificial intelligence module and the third artificial intelligence module, wherein the first artificial intelligence The method may further include generating and storing an artificial intelligence module map connecting a module and the second artificial intelligence module, and connecting the first artificial intelligence module, the second artificial intelligence module, and the third artificial intelligence module. have.

In various embodiments, the generating and storing of the artificial intelligence module map includes storing a new artificial intelligence module collected from the outside, and analyzing the properties of the new artificial intelligence module to have a degree of similarity with a plurality of pre-stored artificial intelligence modules. and linking and storing the new artificial intelligence model with an artificial intelligence module having the determined similarity greater than or equal to a reference value among the plurality of previously stored artificial intelligence modules.

In various embodiments, the providing of the one or more artificial intelligence modules includes: providing the first artificial intelligence module in response to obtaining a request to provide the first artificial intelligence module from the user, based on the artificial intelligence module map It may include providing at least one of the second artificial intelligence module and the third artificial intelligence module connected to the first artificial intelligence module together.

In various embodiments, the providing of the one or more artificial intelligence modules includes: obtaining use information indicating a purpose of use for the first artificial intelligence module together with a request for providing the first artificial intelligence module from the user; The method may include providing an information list on one or more recommended AI modules corresponding to the purpose of use information among a plurality of AI modules connected to the first AI module based on a module map.

In various embodiments, analyzing and processing image data collected from a plurality of camera modules, obtaining a request for recommending one or more artificial intelligence modules applicable to a plurality of image data relay devices that transmit to a server, and the plurality of image data relay devices The method further comprises providing one or more recommended AI models based on each installation environment and installation purpose, wherein the one or more recommended AI models analyze the image data to recognize an object included in the image data, , a first artificial intelligence module for determining the properties of the recognized object, and a second artificial intelligence module for determining whether an event has occurred for the object in the image data and a property of the generated event.

In various embodiments, the providing of the one or more recommended AI models may include, when a second recommended AI module is generated by modifying the first recommended AI module applied to the first image data relay device, the first image The method may include synchronizing a data relay device and a second image data relay device to which the first recommendation AI module is applied to provide the second recommendation AI module to the second image data relay device.

A modularized artificial intelligence model providing service providing apparatus according to another embodiment of the present invention for solving the above-described problems, a memory for storing one or more instructions, and a processor for executing the one or more instructions stored in the memory, The processor may perform the modularized artificial intelligence model providing service providing method according to an embodiment of the present invention by executing the one or more instructions.

A computer program according to another embodiment of the present invention for solving the above-described problems is combined with a computer that is hardware, so that the computer can perform the modularized artificial intelligence model providing service providing method according to an embodiment of the present invention may be stored in a readable recording medium.

Other specific details of the invention are included in the detailed description and drawings.

According to various embodiments of the present invention, an artificial intelligence module generated by modularizing an artificial intelligence model is provided, but the user directly modifies and combines the artificial intelligence module by providing the authority for the artificial intelligence module together with the artificial intelligence module. By making it usable, there is an advantage in that it is possible to use an artificial intelligence module suitable for the purpose of use at a low cost.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a diagram illustrating a modularized artificial intelligence model providing service providing system according to an embodiment of the present invention.
2 is a hardware configuration diagram of a modularized artificial intelligence model providing service providing apparatus according to another embodiment of the present invention.
3 is a flowchart of a method for providing a modularized artificial intelligence model providing service according to another embodiment of the present invention.
4 is a diagram exemplarily illustrating an artificial intelligence module map generated by a modularized artificial intelligence model providing service providing apparatus according to various embodiments.
5 and 6 are diagrams exemplarily illustrating a configuration of adding a new artificial intelligence module generated by modifying an artificial intelligence module to an artificial intelligence module map, in various embodiments.
7 is a diagram illustrating an image data relay system to which a modularized artificial intelligence model providing service providing system is applicable, according to various embodiments.

Advantages and features of the present invention and methods of achieving them will become apparent 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 allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

As used herein, the term “unit” or “module” refers to a hardware component such as software, FPGA, or ASIC, and “unit” or “module” performs certain roles. However, “part” or “module” is not meant to be limited to software or hardware. A “unit” or “module” may be configured to reside on an addressable storage medium or to reproduce one or more processors. Thus, as an example, “part” or “module” refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Components and functionality provided within “parts” or “modules” may be combined into a smaller number of components and “parts” or “modules” or as additional components and “parts” or “modules”. can be further separated.

Spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe the correlation between a component and other components. Spatially relative terms should be understood as terms including different directions of components during use or operation in addition to the directions shown in the drawings. For example, when a component shown in the drawing is turned over, a component described as “beneath” or “beneath” of another component may be placed “above” of the other component. can Accordingly, the exemplary term “below” may include both directions below and above. Components may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

In this specification, a computer refers to all types of hardware devices including at least one processor, and may be understood as encompassing software configurations operating in the corresponding hardware device according to embodiments. For example, a computer may be understood to include, but is not limited to, smart phones, tablet PCs, desktops, notebooks, and user clients and applications running on each device.

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

Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and at least a portion of each step may be performed in different devices according to embodiments.

1 is a diagram illustrating a modularized artificial intelligence model providing service providing system according to an embodiment of the present invention.

Referring to FIG. 1 , a modularized artificial intelligence model providing service providing system according to an embodiment of the present invention includes a modularized artificial intelligence model providing service providing apparatus 100 , a user terminal 200 and an external server 300 . can do.

Here, the modularized artificial intelligence model providing service providing system shown in FIG. 1 is according to an embodiment, and its components are not limited to the embodiment shown in FIG. 1, and may be added, changed, or deleted as necessary. can

In an embodiment, the modularized artificial intelligence model providing service providing apparatus 100 may store a plurality of artificial intelligence modules generated by modularity of a plurality of artificial intelligence models, and according to a user's request, a plurality of artificial intelligence An artificial intelligence module providing service that provides one or more artificial intelligence modules among the modules may be provided.

Here, the provision of the artificial intelligence module may mean that the artificial intelligence module itself is provided to an external device so that the right to use the artificial intelligence module is provided or the artificial intelligence module can be directly installed and applied to the external device. not limited

In various embodiments, the modularized artificial intelligence model providing service providing apparatus 100 may recommend and provide an artificial intelligence module optimized according to the user's purpose of use through the artificial intelligence module providing service. For example, when a user requests an artificial intelligence module capable of performing a function of detecting whether an event has occurred from image data for the purpose of crime prevention, a plurality of artificial intelligence modules stored in connection with the artificial intelligence module for detecting whether an event has occurred Among the intelligence modules, an artificial intelligence module that detects crime-related events can be selected and recommended. However, the present invention is not limited thereto.

In various embodiments, the modularized artificial intelligence model providing service providing apparatus 100 may improve one or more artificial intelligence modules provided to the user, such as modifying, combining, etc., or developing a new artificial intelligence module based on the artificial intelligence module. platform can be provided.

In one embodiment, the user terminal 200 may be connected to the apparatus 100 for providing a service providing a modular artificial intelligence model through the network 400, and the apparatus for providing a service providing a modular artificial intelligence model through the network 400 ( 100) can provide artificial intelligence module provision service.

In various embodiments, the user terminal 200 may include a smart phone, a tablet PC, a desktop and a laptop having a display on at least a portion of the user terminal 200, and provides a modularized artificial intelligence model providing service through the display A user interface (UI) provided by the device 100 (eg, an artificial intelligence module providing UI, an artificial intelligence module development UI, etc.) may be output. However, the present invention is not limited thereto.

In one embodiment, the external server 300 may be connected to the modularized artificial intelligence model providing service providing apparatus 100 through the network 400, and the modularized artificial intelligence model providing service providing apparatus 100 is the artificial intelligence module. It is possible to store and manage various information necessary to provide the provided service. For example, the external server 300 may store and manage data related to a plurality of artificial intelligence modules and data for a plurality of users subscribed to an artificial intelligence module providing service.

In various embodiments, the modularized artificial intelligence model providing service providing apparatus 100 does not include a separate external server 300 and provides a module by a storage device provided in the modularized artificial intelligence model providing service providing apparatus 100 . Various information necessary to provide services may be stored and managed. However, the present invention is not limited thereto. Hereinafter, a hardware configuration of the modularized artificial intelligence model providing service providing apparatus 100 will be described with reference to FIG. 2 .

2 is a hardware configuration diagram of a modularized artificial intelligence model providing service providing apparatus according to another embodiment of the present invention.

Referring to FIG. 2 , an apparatus 100 for providing a modular artificial intelligence model (hereinafter, "computing apparatus 100") according to another embodiment of the present invention may include a processor 110 and a memory 120 . . In various embodiments, the computing device 100 may further include a network interface (or communication interface) (not shown), storage (not shown), and a bus (not shown).

In an embodiment, the processor 110 may control the overall operation of each component of the computing device 100 . The processor 110 may include a central processing unit (CPU), a micro processor unit (MPU), a micro controller unit (MCU), or any type of processor well known in the art.

In various embodiments, the processor 110 may perform an operation on at least one application or program for executing the method according to the embodiments of the present invention. In various embodiments, the processor 110 includes one or more cores (not shown) and a graphic processing unit (not shown) and/or a connection path (eg, a bus, etc.) for transmitting and receiving signals to and from other components. can do.

In various embodiments, the processor 110 temporarily and/or permanently stores a signal (or data) processed inside the processor 110 , a random access memory (RAM) and a read access memory (ROM). -Only Memory, not shown) may be further included. In addition, the processor 110 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, a RAM, and a ROM.

In one embodiment, the processor 110 executes one or more instructions stored in the memory 120 to perform a method (eg, a method for providing a modularized artificial intelligence model providing service) to be described in relation to FIG. 3 . can

For example, the processor 110 executes one or more instructions stored in the memory 120, thereby obtaining a plurality of artificial intelligence modules generated by modularity of each of the plurality of artificial intelligence models, and the artificial intelligence module from the user. It is possible to perform a modularized artificial intelligence model providing service providing method including the step of providing one or more artificial intelligence modules corresponding to the artificial intelligence module provision request among a plurality of artificial intelligence modules in response to obtaining the provision request. However, the present invention is not limited thereto.

In one embodiment, the memory 120 may store various data, commands, and/or information. The memory 120 may store programs (one or more instructions) for processing and controlling the processor 110 . Programs stored in the memory 120 may be divided into a plurality of modules according to functions.

In various embodiments, steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

The components of the present invention may be implemented as a program (or application) to be executed in combination with a computer, which is hardware, and stored in a medium. Components of the present invention may be implemented as software programming or software components, and similarly, embodiments may include various algorithms implemented as data structures, processes, routines, or combinations of other programming constructs, including C, C++ , may be implemented in a programming or scripting language such as Java, assembler, or the like. Functional aspects may be implemented in an algorithm running on one or more processors. Hereinafter, a method of providing a modularized artificial intelligence model providing service performed by the computing device 100 will be described with reference to FIG. 3 .

3 is a flowchart of a method for providing a modularized artificial intelligence model providing service according to another embodiment of the present invention.

Referring to FIG. 3 , in step S110 , the computing device 100 may collect a plurality of artificial intelligence modules generated by modularity of each of a plurality of artificial intelligence models. For example, the computing device 100 may be connected to a terminal of a developer who develops an artificial intelligence model (eg, a device for developing and generating an artificial intelligence model) to collect a plurality of artificial intelligence modules from the terminal of the developer. .

Here, the artificial intelligence module may be a module generated by modularizing an artificial intelligence model learned by a plurality of learning data in order to perform a specific function.

In various embodiments, the computing device 100 may provide a separate artificial intelligence module development platform capable of developing an artificial intelligence module with the user terminal 200 , and an artificial intelligence module developed through the artificial intelligence module development platform can be collected

In various embodiments, the computing device 100 may collect a plurality of artificial intelligence modules (all or some) from a server (eg, the external server 300) in which the plurality of artificial intelligence modules are pre-stored.

In various embodiments, the computing device 100 may collect a plurality of artificial intelligence models from the outside, and may generate a plurality of artificial intelligence modules by performing modularization on the collected artificial intelligence models. However, the present invention is not limited thereto, and any method in which the computing device 100 collects a plurality of artificial intelligence modules may be applied.

In step S120, the computing device 100 may store the plurality of artificial intelligence modules collected in step S110.

In various embodiments, the computing device 100 may store a plurality of artificial intelligence modules in the form of an artificial intelligence module map indicating a connection relationship with a plurality of artificial intelligence modules collected through step S110.

For example, the computing device 100 may include a second artificial intelligence module and a first artificial intelligence generated by modifying the first artificial intelligence module from the user in response to the first artificial intelligence module provided to the user through step S130 to be described later. obtain a third artificial intelligence module generated by combining at least a part of the module and the second artificial intelligence module, and store the second artificial intelligence module and the third artificial intelligence module, the first artificial intelligence module and the second artificial intelligence module It is possible to create and store an artificial intelligence module map linking the first artificial intelligence module, the second artificial intelligence module, and the third artificial intelligence module. Hereinafter, with reference to FIGS. 4 to 6 , a method for the computing device 100 to generate an artificial intelligence module map and a form of the artificial intelligence module map generated accordingly will be described.

4 is a diagram exemplarily illustrating an artificial intelligence module map generated by a modularized artificial intelligence model providing service providing apparatus in various embodiments, and FIGS. 5 and 6 are, in various embodiments, as the artificial intelligence module is modified. It is a diagram exemplarily showing a configuration for adding a new artificial intelligence module to be created to an artificial intelligence module map.

4 to 6 , the computing device 100 stores a plurality of artificial intelligence modules, and may generate an artificial intelligence module map by connecting a plurality of artificial intelligence modules according to the respective correlations of the plurality of artificial intelligence modules. have.

First, when the computing device 100 is newly created based on the artificial intelligence module collected according to the artificial intelligence module collection operation (eg, step S110 in FIG. 3 ), the collected artificial intelligence module and the newly created You can create an AI module map that connects AI modules.

For example, as shown in FIG. 5 , when a second artificial intelligence module is newly created based on the first artificial intelligence module, the computing device 100 (eg, a plurality of unit modules included in the first artificial intelligence module) (when a new artificial intelligence module is created by modifying at least one unit module among

At this time, the computing device 100 connects the second artificial intelligence module and the first artificial intelligence module, respectively, to identify that the second artificial intelligence module is generated based on the first artificial intelligence module. The module can be connected with a line in the form of an arrow in the direction of the second artificial intelligence module.

In addition, when the computing device 100 is newly created on the basis of two or more artificial intelligence modules among a plurality of artificial intelligence modules previously stored in the artificial intelligence module map, the two or more artificial intelligence modules and the newly created artificial intelligence You can create an AI module map that connects modules.

For example, as shown in FIG. 6 , the computing device 100 generates a new fourth artificial intelligence module based on the second artificial intelligence model and the third artificial intelligence model (eg, the second artificial intelligence module). When it is newly created by combining at least some modules of the third artificial intelligence model and at least some modules of can do.

In various embodiments, the computing device 100 connects the first artificial intelligence module and the second artificial intelligence module derived from the first artificial intelligence module, but information ( Example: information on which parts of the first artificial intelligence model were supplemented, corrected, added, and deleted) can be provided together.

In various embodiments, the computing device 100 includes a new artificial intelligence module (eg, an artificial intelligence module not derived by a pre-stored artificial intelligence module) in a state in which a plurality of artificial intelligence modules are pre-stored in the form of an artificial intelligence module map. is collected, the similarity with a plurality of pre-stored AI modules is determined by analyzing the properties of the new AI module, and the determined similarity among the plurality of pre-stored AI modules is greater than or equal to the reference value and the new AI module and the new AI module. Models can be linked and saved.

For example, as shown in FIG. 6 , in the computing device 100 , among a plurality of artificial intelligence modules, an artificial intelligence module having a similarity with a new artificial intelligence module equal to or greater than a reference value is a first artificial intelligence module and a second artificial intelligence module, respectively. In the case of a module, the first artificial intelligence module, the second artificial intelligence module, and the new artificial intelligence module may be connected by a single line, respectively. At this time, since the new artificial intelligence module is not an artificial intelligence module derived from the first artificial intelligence module and the second artificial intelligence module, the computing device 100 includes the first artificial intelligence module and the second artificial intelligence module and the new artificial intelligence module. It can be connected with lines, not arrows.

In various embodiments, the computing device 100 provides information related to the similarity between the new AI module and the new AI module and the similarity between the new AI module and the new AI module on an area adjacent to one line connected to the new AI module. can

Again, referring to FIG. 3, in step S130, the computing device 100 provides one or more artificial intelligence modules corresponding to the artificial intelligence module provision request among the plurality of artificial intelligence modules in response to obtaining the artificial intelligence module provision request from the user. can do.

For example, the computing device 100 may provide an AI module providing service UI to the user terminal 200 , and may receive one or more keywords indicating functions of the AI module from the user through the UI, It is possible to provide an artificial intelligence module list including information on a plurality of artificial intelligence modules corresponding to the above keywords.

Thereafter, the computing device 100 may obtain an artificial intelligence module provision request for selecting any one of a plurality of artificial intelligence modules included in the artificial intelligence module list from the user through the UI, and one corresponding to the artificial intelligence module provision request More than one artificial intelligence module can be provided.

Here, the provision of the artificial intelligence module may mean that the artificial intelligence module itself is provided to an external device so that the right to use the artificial intelligence module is provided or the artificial intelligence module can be directly installed and applied to the external device. not limited

In various embodiments, the computing device 100 selectively extracts and provides only modules corresponding to one or more keywords among a plurality of unit modules included in a plurality of artificial intelligence modules based on one or more keywords input from the user, By combining the unit modules provided by the user, it can be facilitated to create a new artificial intelligence module.

In various embodiments, the computing device 100 provides the first artificial intelligence module in response to obtaining a request to provide the first artificial intelligence module from the user, wherein the artificial intelligence connected to the first artificial intelligence module based on the artificial intelligence module map If there is a module, an artificial intelligence module connected to the first artificial intelligence module may be additionally provided together.

For example, as shown in FIG. 4 , the computing device 100 includes a first artificial intelligence module, a second artificial intelligence module, a third artificial intelligence module, a fourth artificial intelligence module, a fifth artificial intelligence module, and a sixth artificial intelligence module. When the intelligence module is connected, the second to sixth artificial intelligence modules may be provided together with the first artificial intelligence module in response to obtaining a request to provide the first artificial intelligence module from the user. However, the present invention is not limited thereto, and the computing device 100 may provide only the first artificial intelligence module, but may recommend and provide information on the second to sixth artificial intelligence modules connected to the first artificial intelligence module.

In various embodiments, the computing device 100 may obtain usage information indicating the purpose of use for the first artificial intelligence module together with a request for providing the first artificial intelligence module from the user, and based on the artificial intelligence module map, the first artificial intelligence module It is possible to provide an information list on one or more recommended artificial intelligence modules corresponding to the purpose of use among a plurality of artificial intelligence modules connected to the intelligence module.

For example, the computing device 100 analyzes image data (or voice data or IoT sensor data) from the user and uses the artificial intelligence module together with the artificial intelligence module to detect an event for the purpose of preventing crime. In the case of obtaining , it is possible to recommend and provide an artificial intelligence module that analyzes an object from image data among a plurality of artificial intelligence modules connected to an artificial intelligence model that detects an event by analyzing image data to detect a crime-related event occurring in the object. have.

In various embodiments, the computing device 100 may provide the artificial intelligence module requested by the user, but may also provide the recommended artificial intelligence module according to the properties of the device to which the artificial intelligence module is applied. Hereinafter, it will be described with reference to FIG.

7 is a diagram illustrating an image data relay system to which a modularized artificial intelligence model providing service providing system is applicable, according to various embodiments.

Referring to FIG. 7 , the computing device 100 analyzes and processes image data collected from a plurality of camera modules 22 and transmits the image data relay devices 21-1 to 21-N to the server 23 . ), one or more AI module recommendation requests applicable to the device can be obtained, and one or more recommended AI models are selected based on the installation environment and installation purpose of each of the plurality of image data relay devices 21-1 to 21-N. can provide

Here, the video data relay device 21 can perform a function of overall controlling the CCTV system of a certain scale (eg, one factory and a building unit), and divides a predetermined area into an area of a certain size, A plurality of image data relay devices 21-1 to 21-N may be included to control the CCTV system for each area.

In addition, here, the one or more recommended AI models analyze the image data to recognize an object included in the image data, and a first artificial intelligence module (eg, Darknet-based object recognition model) for determining the properties of the recognized object; and It may include, but is not limited to, a second artificial intelligence module (eg, an event detection model such as motion detection) for determining whether an event has occurred for an object in the image data and a property of the generated event.

For example, the computing device 100 recommends and provides an artificial intelligence model for detecting a fire event with respect to a plurality of image data relay devices 21-1 to 21-N applied to a CCTV system in an area where fires frequently occur. It is possible to recommend an artificial intelligence model that detects intrusion and theft events for a plurality of video data relay devices (21-1 to 21-N) applied to CCTV systems in areas where valuables are stored, such as banks and museums. .

In various embodiments, when the second recommendation AI module is generated by modifying the first recommendation AI module applied to the first image data relay device, the computing device 100 performs the first image data relay device and the first recommendation AI module. The second recommended artificial intelligence module may be provided to the second image data relay device by synchronizing the second image data relay device to which the intelligence module is applied.

For example, the computing device 100 is an artificial intelligence module for analyzing an event in which a person falls applied to the first image data relay device is improved, and an artificial intelligence module for determining whether a person is injured by a fall is generated. In this case, in addition to the first image data relay device, an artificial intelligence module that analyzes an event in which a person falls by synchronizing a plurality of image data relay devices to which an artificial intelligence module for analyzing an event of a person falls is applied and the first image data relay device An artificial intelligence module for determining whether a person is injured by a fall may be provided to each of the plurality of image data relay devices to which this has been applied. However, the present invention is not limited thereto.

The computing device 100 and the server 23 store modularized artificial intelligence models according to the disclosed embodiment, and the amount of data transferred between the computing device 100 and the server 23 can be adjusted by using them.

For example, the computing device 100 generates a plurality of unit image data by dividing each of one or more image data in which an object is recognized among a plurality of image data, and one corresponding to a region including the object among the plurality of unit image data. More than one unit image data can be selected.

Thereafter, the computing device 100 may generate object image data (a form in which unit image data corresponding to a region not including an object is deleted) by cutting only one or more unit image data corresponding to the region including the object.

As another example, the computing device 100 generates a plurality of unit image data by dividing each of one or more image data in which an object is recognized among a plurality of image data, and corresponds to a region that does not include an object among the plurality of unit image data. One or more unit image data may be selected.

Thereafter, the computing device 100 corrects one or more unit image data corresponding to a region not including an object with respect to each of the one or more image data in a direction in which the capacity of the image data decreases (eg, lowering the resolution and changing the screen to an achromatic color). configuration, number of frames, etc.)

For example, the computing device 100 may adjust the number of frames according to the importance of image data. For example, in the image of the part where the object is recognized, the number of frames may not be reduced in order to determine the motion of the object, or a relatively large number of frames may be transmitted to the server. On the other hand, the image of the part in which the object is confirmed not to be recognized can be transmitted to the server by reducing the number of frames relatively significantly.

In an embodiment, the server may perform object recognition on the image transmitted from the computing device 100 again. In the process, an object that the computing device 100 does not recognize or an artificial intelligence model stored in the computing device 100 does not recognize another type of object (eg, a model that recognizes a person in the computing device 100) However, when a model for recognizing an animal or a specific object may not be mounted) is recognized, an image of a predetermined time range before and after the frame in which the object is recognized is retransmitted to the computing device 100 (eg, more frames or higher resolution, etc.).

In an embodiment, the computing device 100 may adjust the number of frames transmitted to the server according to the movement of the recognized object. For example, the computing device 100 may transmit a larger number of frames to the server as the change between frames increases, but is not limited thereto.

In an embodiment, the computing device 100 may transmit predetermined information to the server based on an interval between transmitted frames. For example, code information based on the number of skipped frames may be transmitted to the server, thereby reducing the amount of transmitted data. As a specific example, if a frame is sampled at intervals of 5 frames in the image, there is no recognized object, if frames are sampled at intervals of 3 frames, a relatively low-importance object is recognized, and when frames are sampled at intervals of 1 frame Assuming that an object of relatively high importance is recognized, the server may determine the type of the object included in the corresponding frames only with the interval between the transmitted frame numbers, but is not limited thereto. In addition, more information may be transmitted based on the interval of several frames. For example, when an image sampled at intervals of one frame three times in a row is received, the server may determine it as a header, and may acquire information based on the frame interval of an image transmitted thereafter. For example, when frames sampled at 3 frame intervals, 2 frame intervals, and 5 frame intervals are received after the header, it may be determined that the object of the code corresponding to 325 is included in the corresponding image. A code table for this may be shared between the server and the computing device in advance.

The above-described modularized artificial intelligence model providing service providing method has been described with reference to the flowchart shown in the drawings. For a simple description, the modularized artificial intelligence model providing service providing method has been described by showing a series of blocks, but the present invention is not limited to the order of the blocks, and some blocks are shown in an order different from that shown and operated in this specification. may be performed or may be performed simultaneously. In addition, new blocks not described in the present specification and drawings may be added, or some blocks may be deleted or changed.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains know that the present invention may be embodied in other specific forms without changing the technical spirit or essential features thereof. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

100: modularized artificial intelligence model providing service providing device (or computing device)
200: user terminal
300: external server
400: network

Claims (10)

A method performed by a computing device, comprising:
obtaining a plurality of artificial intelligence modules generated by modularity of each of a plurality of artificial intelligence models; and
Comprising the step of providing one or more artificial intelligence modules corresponding to the artificial intelligence module provision request among the plurality of artificial intelligence modules in response to obtaining an artificial intelligence module provision request from a user,
A method of providing modularized artificial intelligence model platform services. According to claim 1,
The step of providing the one or more artificial intelligence modules comprises:
providing authorization to the one or more artificial intelligence modules together with the one or more artificial intelligence modules;
A method of providing modularized artificial intelligence model platform services. According to claim 1,
The step of obtaining the plurality of artificial intelligence modules comprises:
Combining at least a part of a second artificial intelligence module generated by modifying the first artificial intelligence module from the user in response to the first artificial intelligence module provided to the user, and the first artificial intelligence module and the second artificial intelligence module obtaining a third artificial intelligence module generated by and
store the second artificial intelligence module and the third artificial intelligence module, wherein the first artificial intelligence module and the second artificial intelligence module are connected, the first artificial intelligence module and the second artificial intelligence module and the second artificial intelligence module 3 further comprising the step of creating and storing an artificial intelligence module map that connects the three artificial intelligence modules,
A method of providing modularized artificial intelligence model platform services. 4. The method of claim 3,
The step of generating and storing the artificial intelligence module map comprises:
Storing a new artificial intelligence module collected from the outside, analyzing the properties of the new artificial intelligence module to determine a degree of similarity with a plurality of pre-stored artificial intelligence modules, and determining the degree of similarity among the plurality of pre-stored artificial intelligence modules Comprising the step of connecting and storing the new artificial intelligence model with an artificial intelligence module that is equal to or greater than a reference value,
A method of providing modularized artificial intelligence model platform services. 4. The method of claim 3,
The step of providing the one or more artificial intelligence modules comprises:
providing the first artificial intelligence module in response to obtaining a request to provide a first artificial intelligence module from the user, wherein the second artificial intelligence module and the second artificial intelligence module connected to the first artificial intelligence module based on the artificial intelligence module map 3 comprising the step of providing together at least one of the artificial intelligence modules,
How to provide modularized AI model platform service. 4. The method of claim 3,
The step of providing the one or more artificial intelligence modules comprises:
obtaining usage information indicating a purpose of use for the first AI module together with a request for providing a first AI module from the user; and
Comprising the step of providing an information list on one or more recommended artificial intelligence modules corresponding to the use purpose information among a plurality of artificial intelligence modules connected to the first artificial intelligence module based on the artificial intelligence module map,
How to provide modularized AI model platform service. According to claim 1,
obtaining a request for recommending one or more artificial intelligence modules applicable to a plurality of image data relay devices that analyze and process image data collected from a plurality of camera modules and transmit them to a server; and
Further comprising the step of providing one or more recommended AI models based on the installation environment and installation purpose of each of the plurality of image data relay devices,
The one or more recommended artificial intelligence models are
A first artificial intelligence module that analyzes the image data to recognize an object included in the image data, and determines a property of the recognized object, and whether an event has occurred for the object in the image data and a property of the generated event Containing a second artificial intelligence module to determine,
How to provide modularized AI model platform service. 8. The method of claim 7,
The step of providing the one or more recommended artificial intelligence models comprises:
When the second recommended AI module is generated by modifying the first recommended AI module applied to the first image data relay device, relaying the first image data relay device and the second image data to which the first recommended AI module is applied and providing the second recommended artificial intelligence module to the second image data relay device by synchronizing a device,
How to provide modularized artificial intelligence model platform service. a memory storing one or more instructions; and
a processor executing the one or more instructions stored in the memory;
The processor by executing the one or more instructions,
An apparatus for performing the method of claim 1 . A computer program stored in a computer-readable recording medium in combination with a computer, which is hardware, to perform the method of claim 1 .

Images