KR101986804B1 - Method and apparatus for database of searching visual contents - Google Patents

Abstract

The present invention relates to an apparatus and method for generating a visual content search database in which a plurality of basic information about a plurality of foods are stored. To this end, a modified image corresponding to any one of basic information items in the visual content search database An input part for receiving an input image and an input part for inputting at least one of the color of the content, the position of the content in the image, the content of the content, and the shape of the container containing the content, An attribute extraction unit configured to extract an attribute of each of the plurality of discrimination attributes and a plurality of discrimination attributes matching each of the basic information and having representative features for mutual discrimination and data values matched to each of the plurality of discrimination attributes; , And using the attribute value extracted by the attribute extracting unit And a search unit configured to generate search comparison information linked to any one of the basic information based on the updated attribute training set and to generate search comparison information based on the data linked to one of the basic information in the database for searching visual contents, And a data generating unit for generating a visual content search database.

Description

[0001] METHOD AND APPARATUS FOR DATABASE OF SEARCHING VISUAL CONTENTS [0002]

The present invention relates to an apparatus and method for generating a database for visual content retrieval.

Numerous multimedia contents are being shared on the Internet (web), but up to now, text search is mainly performed on the web. However, there is not yet a fast and efficient search method for motion video data including image, video, audio data, and voice.

 In recent years, as the amount of multimedia data increases, it is required to construct a database for multimedia data, and to provide a search service to a user by using the constructed database.

Korean Registered Patent No. 10-0754157 (Registered on September 3, 2007)

The present invention relates to a method and apparatus for extracting attribute values based on an image obtained by transforming visual content including any food into various forms and generating search comparison information to be used in any one of food searches using the extracted attribute values An apparatus and method for generating a database for visual content search are provided.

According to another aspect of the present invention, there is provided an apparatus for generating a visual content search database, the apparatus for generating a visual content search database storing a plurality of basic information about a plurality of foods, An input unit for inputting a modified image corresponding to any basic information in the database for visual content retrieval; and an input unit for inputting the color of the content, the content of the content, An attribute extracting unit that extracts at least one of a position in the image, a shape of the content, and a shape of the container containing the content; and a plurality of feature extraction units matching the one basic information and having representative features for separating each other The discrimination attribute and the day matched to each of the plurality of discrimination attributes Wherein the attribute training set is updated by using an attribute set extracted by the attribute extracting unit, and the attribute training set is updated based on the updated attribute training set, And a data generation unit for generating data linked to one of the basic information items in the database for visual content search based on the generated information.

According to an embodiment of the present invention, the database for generating a visual content search may set photographing environment information of a camera for photographing the modified image, or may receive the photographing environment information through interlocking with the camera, The attribute extracting unit may extract the photographing environment information provided through the photographing environment setting unit as an attribute value of the modified image and provide the extracted photographing environment information to the data generating unit.

According to an embodiment of the present invention, the apparatus for generating a visual content search further includes an image storage unit for mapping the modified image and the extracted attribute value using one of the basic information as an index and storing the mapped attribute value in an image database The data generation unit may generate search comparison information for one of the basic information through machine learning using the modified image and the attribute value stored in the image storage unit.

According to another aspect of the present invention, there is provided a method of generating a database for visual content retrieval, the method comprising: The method comprising the steps of: receiving a modified image corresponding to any one of basic information items in the visual content search database; analyzing the modified image to determine a color of the content, Extracting at least one of a position in the image, a shape of the content, and a shape of the container containing the content; extracting at least one of a position, A plurality of classifying attributes each having a plurality of classifying attributes May comprise the steps of using the updated attribute training set produced the search information comparison for coupled to the one of the basic information to update the properties training set consisting of data values matched to each.

According to an embodiment of the present invention, the method for creating a database for visual content search further includes setting shooting environment information of a camera for shooting the modified image or receiving photographing environment information through interlocking with the camera And the extracting step may extract the setting environment information, which is set or provided, as an attribute value of the modified image.

According to an embodiment of the present invention, the method of generating a database for visual content retrieval includes the steps of mapping the modified image and the extracted attribute values using one of the basic information as an index and storing the mapped attribute value in an image database, The generating step may generate search comparison information for any one of the basic information through machine learning using the stored modified image and attribute values.

According to an embodiment of the present invention, an attribute value is extracted based on an image obtained by transforming a visual content including a certain food into various forms, By providing an apparatus and method for creating a visual content search database capable of generating comparison information, it is possible to improve the retrieval accuracy of various food related visual contents.

BRIEF DESCRIPTION OF DRAWINGS FIG. 1 is an overall network configuration diagram illustrating a visual content search system according to an embodiment of the present invention; FIG.
2 is a block diagram showing a detailed configuration of a database for generating a visual content search according to an embodiment of the present invention.
3 is a flowchart illustrating a process of building an image database by a database for creating a visual content search database according to an exemplary embodiment of the present invention.
4 is a flowchart illustrating a process of generating search comparison information based on data stored in an image database according to an embodiment of the present invention.
5 is a block diagram illustrating a detailed configuration of a visual content search request processing apparatus according to an embodiment of the present invention.
6 is a diagram illustrating an interface to a visual browser displayed on a display unit of a device equipped with a visual content search request processing apparatus according to an embodiment of the present invention.
7 is a flowchart illustrating a search process of visual contents according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification of the present invention, when a part is referred to as "including " an element, it is understood that it may include other elements as well, without excluding other elements unless specifically stated otherwise.

The terms "about "," substantially ", etc. used to the extent that they are used throughout the present disclosure are used in their numerical value or in close proximity to their numerical values when the manufacturing and material tolerances inherent in the stated meanings are presented, Accurate or absolute numbers are used to prevent unauthorized exploitation by unauthorized intruders of the referenced disclosure. The word " step (or step) "or" step "used in the specification of the present invention does not mean" step for.

In this specification, the term " part " includes a unit realized by hardware, a unit realized by software, and a unit realized by using both. Further, one unit may be implemented using two or more hardware, or two or more units may be implemented by one hardware. In the present description, some of the operations or functions described as being performed by a terminal, a device, or a device may be performed instead in a server connected to the terminal, device, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed in a terminal, device or device connected to the server. Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

BRIEF DESCRIPTION OF DRAWINGS FIG. 1 is an overall network configuration diagram illustrating a visual content search system according to an embodiment of the present invention; FIG.

1, the visual content search system includes a database 100 for visual content search, an image database 110, a search database 120, a visual content search request processing device 130, a search engine (not shown) 140, a content database 150, a recipe database 160, and the like.

First, the visual content search database generating apparatus 100 of FIG. 1 may be connected to the image database 110 and the search database 120 by wire or wireless. In addition, the database 100 for visual content retrieval may be stored in a recording medium in a form executable by at least one or more processors. Specifically, the database 100 for generating a visual content search is executed by at least one processor as an image, for example, an image corresponding to food-related visual content is acquired, extracts attributes of the image, Attribute and photographing environment information are mapped and stored in the image database 110 and machine learning is performed on the basis of the data stored in the image database 110 so as to correspond to search comparison information using the basic information as an index Data, such as annotation information, may be generated and stored in the search database 120.

In addition, the search database 120 connected to the database 100 for visual-content-search database stores basic information about contents including a plurality of foods into a plurality of categories and stores the basic information in a search matching manner A plurality of search comparison information are linked and stored. Here, the plurality of search comparison information may mean the comparison data used in the search of the visual content, and the basic information may mean the search result data as the data named in each of the contents.

For example, the basic information includes japanese, chanpon, kimbap, pasta, and pizza. The japanese and chanpon are classified and stored in the Chinese food category. The pasta and the pizza are classified and stored in the European category. Can be classified and stored. In addition, the search comparison information connected to each basic information may be an attribute value for the corresponding basic information.

The apparatus 100 for creating a visual content search database according to the embodiment of the present invention stores the image in the image database 110 as described above or updates the search comparison information linked to each basic information in the search database 120 .

The detailed configuration and operation of the database 100 for creating a visual content search as described above will be described with reference to the following drawings.

2 is a block diagram showing a detailed configuration of an apparatus 100 for generating a visual content search database according to an embodiment of the present invention.

2, the apparatus 100 for creating a visual content search database includes an input unit 210, a photographing environment information setting unit 220, an attribute extracting unit 230, an image storing unit 240, (250), and the like.

The input unit 210 may receive a plurality of modified images for the visual content data corresponding to any one of the basic information. Specifically, the input unit 210 displays a plurality of modified images for visual content data corresponding to the jammed surface, such as the color of the content, the position value (coordinate value in the image) of the content (food) in the image, Or the shape of the container including the image of the container. For this, the input unit 210 may be interlocked with the camera 200 for photographing the contents.

Meanwhile, the input unit 210 may receive a plurality of modified images and provide them to the attribute extraction unit 230.

The photographing environment information setting unit 220 may provide an interface through which the photographing environment information can be set, or may receive the photographing environment information from the camera 200 through interlocking with the camera 200, The photographing environment information provided from the camera 200 may be provided to the attribute extracting unit 230. [

In the embodiment of the present invention, the photographing environment information may include data related to the operation of the camera 200, such as a zoom state, a flash operation state, and a surrounding environment state (data such as weather, photographing time, and season).

The attribute extracting unit 230 analyzes attributes of each of the plurality of images input through the input unit 210 and determines attribute values of each of the plurality of images based on the photographing environment information provided through the photographing environment information setting unit 220 Can be extracted. Specifically, the attribute extracting unit 230 separates a container containing the content and contents from the image through the analysis of the image, and determines the color of the content, the position value within the image, the shape of the content, Information such as information can be extracted. The attribute values for each of the extracted images may be provided to the image storage unit 240. [

The image storage unit 240 may map the image and extracted attribute values using the basic information as an index, and store the mapped attribute values in the image database 110.

The data generation unit 250 may receive the attribute values mapped to the images and images stored in the image database 110 and perform machine learning to generate search comparison information for basic information. Specifically, the data generating unit 250 is for performing learning on basic information, and can generate data to be compared with the attribute value of the visual content requested to be searched at the time of searching, that is, search comparison information.

For this, the data generation unit 250 may manage the attribute training set 260 for each basic information. Specifically, the data generation unit 250 may classify each attribute in the image and manage a plurality of classification attributes including a plurality of data values (attribute values) for each basic information. In the embodiment of the present invention, the plurality of classification attributes have a plurality of data values (attribute values for the image), and may have discrimination information to be distinguished from other classification attributes, that is, representative characteristics. Here, the classification information may include the color of the basic information, the position value, the content and the form of the container, and each parameter in the photographing environment information.

In addition, the data generator 250 determines whether a property value of an image stored in the image database 110 is included in a certain classification attribute in the attribute training set 260, and then includes an attribute value in the determined data value in the classification attribute To update the data values in the plurality of fractional attributes.

On the other hand, if the difference (distance) between the data value in the classified attribute and the attribute value determined after determining the attribute value of the image is included in the sorting attribute is greater than the predetermined sorting reference value, The attribute training set 260 can be updated by generating a sub-classification attribute. For example, if the attribute value is red, which is the color of the contents, it is determined that the attribute value of the image of the data generation unit 250 is included in the classification attribute related to the color, and then the data value in the color classification attribute Color system) and an attribute value of the image is greater than a predetermined classification reference value, a red classification attribute having a sub-classification attribute, i.e., an attribute value, related to the color classification attribute can be generated. Accordingly, the color discrimination attribute may be subdivided into a red discrimination attribute so that the attribute training set 260 may be updated.

In the case of the classification attribute related to the position value, the sub-classification attribute having the coordinate range value (coordinate range value of the content in the image) is connected to the data value in the classification information, It is possible to update the data value in the sub-fractional attribute corresponding to the coordinate range determined by using the attribute value after determining which coordinate range is close to the coordinate value.

The data generation unit 250 associates the classification attribute, the sub-classification attribute, the data attribute associated with the classification attribute and the sub-classification attribute corresponding to the attribute value of the image with the basic information, and stores the basic information in the search database 120 Linked search comparison information, for example, annotation information for basic information.

A process of creating the search database 120 by the apparatus 100 for generating a visual content search database having the above-described configuration will be described with reference to FIG.

3 is a flowchart illustrating a process of building an image database 110 by a database 100 for creating a visual content search database according to an exemplary embodiment of the present invention.

3, the input unit 210 first provides a modified image of the visual content corresponding to one basic information through interoperation with the camera 200, and provides the modified image to the attribute extraction unit 230 (S300). Specifically, the input unit 210 receives the visual content including the shape of the content, the color, the shape of the container containing the content, the size, the position of the content, and the like, from the camera 200.

The photographing environment information setting unit 220 provides the photographing environment information provided from the camera 200 to the property extracting unit 230 (S302).

Then, the attribute extracting unit 230 extracts attribute values of the transformed image, for example, the color of the contents in the image, the position value in which the contents are positioned in the image, the shape of the contents, The photographing environment information, and the like are extracted (S304).

Then, the image storage unit 240 updates the image database 110 by generating attribute data and image-mapped data and linking the data with the basic information (S306).

The above-described S300 to S306 are repeatedly performed so that the image database 110 may store basic information to which a plurality of deformed images and data mapped with attribute values are connected.

Thereafter, the database 100 for generating a visual content search performs machine learning based on the data stored in the image database 110 to update the attribute training set 260, and generates search comparison information used in the search can do. This will be described with reference to FIG.

4 is a flowchart illustrating a process of generating search comparison information based on data stored in the image database 110 according to an embodiment of the present invention.

As shown in FIG. 4, the data generator 250 sequentially receives a plurality of images linked to one basic information and data (attribute values) mapped to the basic information in the image database 110 at step S400.

Thereafter, the data generating unit 250 selects a sorting attribute corresponding to the attribute value of the image among the plurality of sorting attributes connected to the basic information in the attribute training set 260 (S402) That is, the distance, is analyzed (S404).

Then, the data generating unit 250 determines whether the distance is equal to or greater than a preset reference value (S406).

If it is determined in step S406 that the value is less than the predetermined value, the data generation unit 250 updates the attribute training set 260 by reflecting the attribute value of the image on the selected attribute as the data value (S408).

As a result of the determination in step S406, if it is determined in step S406 that the image data is abnormal, the data generation unit 250 generates a sub-segmentation attribute that is linked to the selected segmentation attribute using the image attribute value (S410) And updates attribute training set 260 (S412).

The data generating unit 250 repeatedly performs the steps of S400 to S412 as described above to generate a data corresponding to one basic information based on a predetermined number or more, The attribute training set 260 is updated, and the search comparison information, i.e., the search comparison information bound to the basic information is generated using the updated attribute training set 260 (S414).

An apparatus and a method for providing a search result of visual content input by using the search comparison information constructed through the above-described method, that is, a visual content corresponding to food will be described with reference to Figs. 5 to 6. Fig.

Prior to the description of the detailed configuration and operation of the visual content search request processing device 130, the visual content search request processing device 130 is capable of communicating in a wired or wireless manner, For example, a mobile device, a computer, a TV, or the like. In other words, the visual content search request processing device 130 may be an application stored in the recording medium in a form executable by at least one or more processors.

On the other hand, the device including the visual content search request processing device 130 may be connected to the search engine 140 interlocked with the search database 120 of FIG. 1 by wire or wirelessly.

The visual content search request processing device 130 extracts a search attribute value (optionally including photographing environment information) for the visual content, and then requests a search based on the extracted search attribute value. The search result, for example, the visual content You can get kind of food (name).

In addition, the visual content search request processing device 130 extracts unique identification information corresponding to the location information in the beacon signal, and transmits the unique identification information to the visual content search request processing device 130. The visual content search request processing device 130 transmits various types of information, i.e., POI information, Can be provided. Here, the POI information may be restaurant information providing kinds of food in the visual contents or various benefit information (information such as discount, coupons, etc.) provided in the restaurant.

5, the visual content retrieval request processing apparatus 130 includes an image capturing unit 500, a visual content capturing unit 510, a beacon signal receiving unit 520, a feature information extracting unit 530 A search request unit 540, a search result providing unit 550, and a display unit 560.

The image capturing unit 500 may provide the image capturing environment information to the feature information extracting unit 530 during image capturing. Here, the photographing environment information may be the operation state of the camera, which is the image photographing unit 500 at the time of photographing the image, and external condition information (environmental information such as weather, temperature at the time of photographing).

The visual content acquisition unit 510 displays an image including food, which is visual content acquired by the image capturing unit 500, on the display unit 560 through execution of a visual browser stored in the recording medium in an operable form in the device .

The beacon signal receiving unit 520 may receive the beacon signal broadcasted within a predetermined radius by activating the visual content as the acquisition or visual browser is executed, and may provide the received beacon signal to the search request unit 540. [

The beacon signal receiving unit 520 according to the embodiment of the present invention can receive a beacon signal in conjunction with an in-device Bluetooth, an infrared communication module, a Wi-Fi module, and an ultrasonic signal receiving module (e.g., a microphone).

6, the visual browser 600, which is operated in conjunction with the visual content search request processing apparatus 130 according to the embodiment of the present invention, A control interface 610 may be provided.

The control interface 610 provides an operation interface for activating the operation of the beacon signal receiving unit 520 and controls the operation of the beacon signal receiving unit 520 according to a user operation such as a touch of the control interface 610 Can be activated. Accordingly, the beacon signal receiving unit 520 interlocks with the in-device devices (Bluetooth, infrared communication module, WiFi module, and ultrasonic signal receiving module (e.g., microphone)) to receive the beacon signal, A beacon signal having a signal strength of a threshold value or more may be extracted and provided to the search result providing unit 550.

The feature information extracting unit 530 extracts the search attribute values necessary for the search of the visual content using the analysis of the image including the visual content, for example, the food and the photographing environment information provided from the image photographing unit 500, And provides the attribute value to the search requesting unit 540. Specifically, the feature information extracting unit 530 extracts the color of the content (food) in the visual content, the position value of the content in the visual content, the form of the content, the form of the container containing the content, Extracts a search attribute value based on the environment information, and provides the extracted search attribute value to the search request unit 540. Accordingly, the search request unit 540 may send a search attribute value to the search engine 140 to request a search.

In addition, the search requesting unit 540 requests the search request unit 540 to distinguish unique identification information from a sound signal of a high frequency band (a beacon signal received through a microphone of the device) having a predetermined variation pattern of unique identification information corresponding to position information of a particular building The extracted unique identification information and unique information of the device may be included in the search attribute value and transmitted to the search engine 140 to request the search.

The search result providing unit 550 may receive the search result from the search engine 140 and display the search result on the display unit 560. In other words, the search result providing unit 550 may display the search result provided from the search engine 140 on the display unit 560 in cooperation with the visual browser.

The search engine 140 may transmit a search result for the visual content, that is, a search result for determining the food in the visual content, in response to the search request of the search request unit 540. [

Specifically, the search engine 140 determines whether a search attribute value is included in a certain discrimination attribute in the attribute training set (~~ in FIG. 2), and then compares the data value in the discrimination attribute with the search attribute value, Discrimination attribute is included in the sub-classification attribute. If it is determined that the sub-classification attribute is not included, it is determined whether or not the sub-classification attribute is included in the sub-classification attribute through comparison between the data value in the sub- A response to the search request can be transmitted with the basic information, for example, the name of the food as the search result.

On the other hand, the search engine 140 includes a content database 150 that manages contents such as restaurant information, discount information, and coupon information provided in a plurality of contents, for example, specific positions (positions for unique identification information) Can be connected. Accordingly, the search engine 140 searches contents in the contents database 150 based on the unique identification information provided from the search request unit 540, extracts the food related information from the retrieved contents, extracts the extracted food related information A comparison between the search comparison information searched after searching at least one basic information and the search comparison information connected to the at least one basic information in the search database 120 and the search attribute value received from the visual content search request processing device 130 Lt; / RTI >

In addition, the search engine 140 may provide the contents retrieved from the contents database 150 to the visual contents retrieval request processing device 130 based on the unique identification information provided from the visual contents retrieval request processing device 130.

On the other hand, the search engine 140 searches the recipe database 160 for recipe information for cooking the search result based on the search result, that is, the type of food in the visual content, and searches the recipe information To the visual content search request processing device 130.

Although the feature information extracting unit 530 is included in the visual content search request processing device 130 of the device connected to the search engine 140 in a wired or wireless manner according to the embodiment of the present invention, The feature information extracting unit 530 may be included in the search engine 140. [ In this case, the search requesting unit 540 may transmit the visual content obtained by the visual content acquiring unit 510 and the photographing environment information provided from the image photographing unit 500 to the search engine 140 to request a search .

A process of performing a search for visual content through interlocking between the visual content search request processing device 130 having the above-described configuration and the search engine 140 will be described with reference to FIG.

7 is a flowchart illustrating a search process of visual contents according to an embodiment of the present invention.

7, the visual content search request processing device 130 analyzes the visual content including the food image input through the image capturing unit 500, the visual content including the food image received from the image capturing unit 500 (Step S700). The beacon signal received by the beacon signal receiving unit 520 is subjected to analysis of the beacon signal (the sound signal of the variation pattern for the unique identification information having the positional information) Information is extracted (S702).

Then, the visual content search request processing device 130 transmits a search attribute value, unique identification information, and unique information of the device on which the visual content search request processing device 130 operates to the search engine 140 to request a search S704).

Accordingly, the search engine 140 searches at least one content mapped to the unique identification information in the content database 150 (S706), and extracts the food-related content from the searched contents (S708).

Then, the search engine 140 retrieves at least one basic information to be used for the search in the search database 120 based on the food-related content and the search comparison information connected thereto (S710).

Thereafter, the search engine 110 derives a search result indicating which food is the food image in the visual content through comparison between at least one basic information searched for and the search comparison information connected thereto and the search attribute value, To the visual content search request processing device 130 (S712).

The process of deriving the search result by the search engine 140 will be described in detail. The search engine 140 selects any one of the basic attribute informa- tion attributes based on the search attribute value, and determines the difference between the data value and the search attribute value linked to the selected attribute Based on the classification attribute selected. This determination process may be performed for each of the search attribute values to determine which basic information can be included in each of the search attribute values, so that a search result can be derived.

In addition, the search engine 140 searches the recipe database 160 for recipe information necessary for cooking the food corresponding to the food image based on the search result, and searches the contents database 150 for contents, For example, discount information related to a food corresponding to a food image, restaurant information, coupon information, and the like, and provides the extracted content and recipe information to the visual content search request processing device 130 (S714).

It will be understood by those skilled in the art that the foregoing description of the present invention is for illustrative purposes only and that those of ordinary skill in the art can readily understand that various changes and modifications may be made without departing from the spirit or essential characteristics of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

100: Database creation device for visual content retrieval
110: Image database
120: Database for retrieval
130: Visual content retrieval request processing device
140: Search engine
150: Content database
160: Recipe database

Claims (6)

An apparatus for generating a visual content search database storing a plurality of basic information about a plurality of foods,
An input unit for inputting a deformed image corresponding to any basic information in the visual content search database;
A photographing environment setting unit for setting photographing environment information of a camera for photographing the deformed image or receiving and sending the photographing environment information through interlocking with the camera;
Extracting at least one of a color of the content, a position in the image of the content, a shape of the content, and a shape of the container including the content as an attribute value of the deformed image, An attribute extraction unit for extracting photographing environment information provided through the photographing environment setting unit as an attribute value of the modified image;
An attribute training set which is composed of a plurality of discrimination attributes matching each of the basic information and each having representative features for distinguishing each other and data values matched to each of the plurality of discrimination attributes,
The attribute training set is updated by using the attribute value extracted by the attribute extracting unit, and search comparison information linked to any one of the basic information is generated based on the updated attribute training set, And a data generating unit for generating data linked to any basic information in the database for visual content retrieval,
Wherein the data generating unit comprises:
A difference between the extracted attribute value and the data value of the discrimination attribute corresponding to the extracted attribute value is calculated, and when it is determined that the calculated difference is larger than the predetermined discrimination reference value, the discrimination attribute corresponding to the extracted attribute value And a sub-classifying attribute having the extracted attribute value to update the attribute training set.
delete The method according to claim 1,
Wherein the visual content search database generation device comprises:
Further comprising an image storage unit for mapping the modified image and the extracted attribute value using one of the basic information as an index and storing the mapped attribute value in an image database,
Wherein the data generating unit comprises:
And generates search comparison information for any one of the basic information through machine learning using the modified image and attribute values stored in the image storage unit.
A method of generating a database for a visual content search in which a plurality of pieces of basic information about a plurality of foods are stored,
Receiving a transformed image corresponding to any one of basic information items in the visual content search database;
Setting photographing environment information of the camera for photographing the deformed image or receiving photographing environment information through interlocking with the camera;
Extracting at least one of a color of the content, a position in the image of the content, a shape of the content, and a shape of the container including the content as an attribute value of the deformed image, Extracting the setting environment information as the attribute value of the modified image;
Updating the attribute training set consisting of a plurality of classification attributes matching each of the basic information and having representative features for mutual classification and data values matched to each of the plurality of classification attributes using the extracted attribute values Step,
And generating search comparison information linked to any one of the basic information using the updated attribute training set,
Wherein the updating comprises:
Calculating a difference between the extracted attribute value and the data value of the discrimination attribute corresponding to the extracted attribute value,
Generating a sub-classifying attribute associated with a classifying attribute corresponding to the extracted attribute value and having the extracted attribute value to update the attribute training set if it is determined that the calculated difference is greater than a predetermined classifying reference value;
And updating the attribute training set by updating a data value associated with the classification attribute corresponding to the extracted attribute value using the extracted attribute value if the calculated difference is less than or equal to a predetermined classification reference value A method for creating a database for retrieving visual content.
delete 5. The method of claim 4,
The method for creating a visual content search database,
Mapping the transformed image and the extracted attribute value using one of the basic information as an index, and storing the transformed image and the extracted attribute value in an image database;
Wherein the generating comprises:
And generating search comparison information on any one of the basic information through machine learning using the stored transformed images and attribute values.

Images