KR102074654B1 - Method, Apparatus and System for Editing Shopping Mall Webpage - Google Patents

Abstract

A method, apparatus, and system for authoring a shopping mall web page are disclosed. In the method of authoring a web page, an image classification engine is trained to classify each input image into a corresponding primary class by inputting a plurality of images into an image classification engine, and related to products related to a product to be sold through a web page. Input images, classify the input product-related images by primary class using the trained image classification engine, classify the product-related images classified by primary class by secondary class based on the correspondence table, Product-related images classified by the second class may be inserted in the corresponding positions of the web page. Therefore, web pages can be authored quickly and easily.

Description

Method, Apparatus and System for Authoring Shopping Mall Webpages {Method, Apparatus and System for Editing Shopping Mall Webpage}

The present invention relates to a method, apparatus, and system for creating a shopping mall web page, and more particularly, a shopping mall web page that facilitates the production of a shopping mall web page based on the web by easily classifying images such as photographs used in the shopping mall. A method and system for authoring.

In general, a shopping mall operator must perform a considerable number of processes to register a product in an internet shopping mall. For example, in order to generate a shopping mall page, a shopping mall operator must photograph various product photos and then upload them to a desired location by sorting them one by one while looking at the product photos according to the shopping mall web page structure that he or she envisions. In addition, a representative image that can induce product sales should be selected and registered as a representative image.

However, such a conventional shopping mall manufacturing method takes a very long time for classifying photos, selecting a photo display position, uploading, and selecting a representative image, and there is also a problem that excessive stress is given to a shopping mall operator for uploading a product sales web page. . In particular, such a problem is further maximized for a seller who sells a large number of items, and there is a problem in that time, mental and financial consumption are severe.

Patent Publication No. 20030010785

An object of the present invention is to provide a shopping mall webpage authoring method, apparatus, and system that can automatically classify an image when a shopping mall webpage is manufactured and arrange it at a designated location.

In order to achieve the above object, the present invention provides a shopping mall web page authoring method in one aspect. The shopping mall authoring method is a method performed by a shopping mall webpage authoring apparatus, comprising: training an image classification engine to input a plurality of images into an image classification engine and classify each input image into a corresponding primary class ; Inputting product related images related to a product to be sold through a web page; Classifying the input product-related images by primary class using the educated image classification engine; Classifying the product-related images classified by the primary class by the secondary class based on a corresponding table; And inserting the product related images classified by the second class into corresponding positions of the web page.

The training of the image classification engine may include a display unit for displaying the received images, a selection unit for selecting one primary class among a plurality of primary classes, and a selection unit selected from the images displayed on the display unit. Designating an image corresponding to a primary class and displaying a user interface that can be input to the image classification engine; The method may include inputting the plurality of images to the image classification engine for each primary class based on the user interface.

The plurality of images may be clothes-related images collected from a plurality of shopping malls. The training of the image classification engine may include inputting the clothing product-related images into the image classification engine until the clothing product-related images for each shopping mall are equalized within a set number of ranges.

The shopping mall web page authoring method may include: a user interface indicating at least one of information indicating how many garment product-related images are input to an image classification engine for each shopping mall and information for comparing the number of image items related to clothing items per mall; The method may further include displaying.

The shopping mall webpage authoring method may further include setting at least one primary class corresponding to each secondary class. Each of the secondary classes may be a group of product related images arranged at a predetermined position on the web page.

The shopping mall web page authoring method includes a display unit displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and an authoring tool capable of authoring the webpage on a single screen. The method may further include displaying a web page authoring screen. Here, when the indication icon is selected by the user, classifying by the primary class, classifying by the secondary class, and inserting may be automatically performed.

The web page authoring screen may display, in real time, a preview form of product-related images inserted for each secondary class in the authoring tool at a corresponding arrangement position.

On the other hand, in order to achieve the above object of the present invention, the present invention provides a shopping mall web page authoring device in another aspect. The shopping mall web page authoring apparatus includes: an engine education unit configured to educate an image classification engine to input a plurality of images into an image classification engine and classify each input image into a corresponding primary class; An input unit for inputting product related images related to a product to be sold through a web page; A primary classifying unit which classifies the input image related products by primary classes using the educated image classification engine; A secondary classification unit classifying the product-related images classified by the primary class by the secondary class based on a corresponding table; And an insertion unit for inserting the product-related images classified by the second class into corresponding positions on the authoring screen of the web page.

The engine education unit includes a display unit for displaying the received images and a selection unit for selecting one primary class among a plurality of primary classes, and corresponding to the primary class selected by the selection unit among the images displayed on the display unit. The user interface may be displayed by specifying an image to be input to the image classification engine. The engine education unit may input the plurality of images to the image classification engine for each primary class based on the user interface.

The plurality of images may be clothes-related images collected from a plurality of shopping malls. The engine education unit may input the clothing goods related images to the image classification engine until the clothing goods related images for each shopping mall are equalized within the set number range.

The engine education unit may display a user interface indicating at least one of information indicating how many clothing goods-related images have been input to the image classification engine for each shopping mall and information for comparing the number of clothing goods-related images for each shopping mall. have.

The secondary classifier may set at least one primary class corresponding to each secondary class. Here, each of the secondary classes may be a group of product-related images arranged at a predetermined position on the web page.

The shopping mall webpage authoring apparatus may display a display unit displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and an authoring tool capable of authoring the webpage on a single screen. It may further include a web page authoring unit for displaying a web page authoring screen.

When the instruction icon is selected by the user, the web page authoring unit controls the primary classifying unit, the secondary classifying unit, and the inserting unit to classify the input product-related images by the primary class, and The process of classifying by class and inserting into the corresponding position of the web page can be performed automatically.

The webpage authoring unit may display, in real time, a preview form of product-related images inserted for each secondary class in the authoring tool of the authoring screen of the webpage in a corresponding arrangement position.

On the other hand, in order to achieve the above object of the present invention the present invention provides a shopping mall web page authoring system in another aspect. The shopping mall web page authoring system includes an image classification engine; Training the image classification engine to classify each input image into a corresponding primary class by inputting a plurality of images to the image classification engine, and input product related images related to a product to be sold through a web page; Using the trained image classification engine, classify the input product-related images by primary class, classify the product-related images classified by the primary class by secondary class based on a correspondence table, and It may include a web page authoring device for inserting the product-related images classified by the second class in a corresponding position on the authoring screen of the web page.

The web page authoring apparatus includes a display unit for displaying the received images and a selection unit for selecting one primary class among a plurality of primary classes, and the primary class selected by the selection unit among the images displayed on the display unit. An image corresponding to may be designated to display a user interface that can be input to the image classification engine. The web page authoring apparatus may input the plurality of images to the image classification engine for each primary class based on the user interface.

The plurality of images may be clothes-related images collected from a plurality of shopping malls. The web page authoring apparatus may input the apparel product related images to the image classification engine until the apparel product related images for each shopping mall are equalized within a set number of ranges.

The web page authoring apparatus displays a user interface indicating at least one of information indicating how many clothing goods-related images have been input to an image classification engine for each shopping mall and information for comparing the number of images of clothing goods for each shopping mall. can do.

The webpage authoring apparatus may set at least one primary class corresponding to each secondary class. Here, each of the secondary classes may be a group of product-related images arranged at a predetermined position on the web page.

The web page authoring apparatus includes: a display unit for displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and an authoring tool for authoring the webpage on a single screen. The web page authoring screen can be displayed.

When the instruction icon is selected by the user, the webpage authoring apparatus classifies the input image-related images by the primary class, classifies the secondary class, and inserts the image into the corresponding position of the webpage. You can do this automatically.

The webpage authoring apparatus may display, in real time, a preview form of product-related images inserted for each secondary class in the authoring tool of the authoring screen of the webpage in a corresponding arrangement position.

On the other hand, in order to achieve the above object of the present invention the present invention provides a shopping mall web page authoring device in another aspect. The shopping mall web page authoring apparatus includes: an engine education unit configured to educate an image classification engine to input a plurality of images into an image classification engine and classify each input image into a corresponding primary class; An input unit for inputting product related images related to a product to be sold through a web page; And a display unit for displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and a webpage authoring screen for displaying an authoring tool of the webpage on a single screen. When the indication icon is selected by the user, the input product-related images are classified by the primary class, the product-related images classified by the primary class are classified by the set secondary class, and the secondary And a web page authoring unit for automatically inserting the product-related images classified by class into corresponding positions of the web page.

On the other hand, in order to achieve the above object of the present invention the present invention provides a shopping mall web page authoring system in another aspect. The shopping mall web page authoring system includes an image classification engine; And classifying a plurality of images into an image classification engine to train the image classification engine to classify each input image into a corresponding primary class, and input image-related images related to a product to be sold through a web page. A display unit for displaying the displayed product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, a webpage authoring screen for displaying the authoring tool of the webpage on a single screen, When the indication icon is selected by the user, the input product-related images are classified by the primary class, the product-related images classified by the primary class are classified by the set secondary class, and by the secondary class. The product-related images classified above correspond to the web page It may include a web page authoring device that automatically performs the process of inserting into the device.

As described above, according to the present invention, when the shopping mall web page is produced, the image may be automatically classified and arranged at a designated position. Therefore, it is possible to reduce time and human consumption consumed in the production of shopping mall web pages.

1 is a block diagram showing the configuration of a shopping mall webpage authoring system according to a preferred embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of a shopping mall web page authoring apparatus illustrated in FIG. 1.
FIG. 3 is a flowchart illustrating an operation flow of a shopping mall webpage authoring apparatus illustrated in FIG. 2.
Figure 4 is an exemplary diagram showing the classification of the primary class in accordance with a preferred embodiment of the present invention.
5 to 22 are exemplary views illustrating images for each primary class.
23 is an exemplary diagram illustrating an image input user interface for inputting an image.
24 is an exemplary diagram illustrating a statistical user interface indicating the number of input images for each shopping mall and the number of input images for each primary class.
25 and 26 are diagrams each illustrating a web page authoring screen displayed by a webpage authoring unit, FIG. 25 shows a state in which product-related images for insertion into a webpage are input, and FIG. 26 shows a product-related image. Indicates a classified state.

As the present invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Hereinafter, with reference to the accompanying drawings, it will be described in detail a preferred embodiment of the present invention. In the following description of the present invention, the same reference numerals are used for the same elements in the drawings and redundant descriptions of the same elements will be omitted.

1 is a block diagram showing the configuration of a shopping mall webpage authoring system according to a preferred embodiment of the present invention.

As shown in FIG. 1, the shopping mall webpage authoring system 1 according to a preferred embodiment of the present invention may interwork with a shopping mall SM through a communication network such as the Internet. The shopping mall SM may mean a terminal for shopping goods that sell products through the Internet. The shopping mall web page production system 1 may interwork with a plurality of shopping malls SM1,..., SMn.

The shopping mall webpage authoring system 1 may include an image classification engine 100 and a shopping mall webpage authoring apparatus 200. The image classification engine 100 and the shopping mall webpage authoring apparatus 200 may interoperate with each other. Here, the image classification engine 100 may be a machine learning or deep learning based artificial intelligence engine trained by the shopping mall webpage authoring apparatus 200.

 The shopping mall webpage authoring apparatus 200 may be implemented based on at least one server-class computer terminal. The computer terminal may include an input device, a display device, a networking device, a hard disk, a memory for storing a program, and a processor for executing a program stored in the memory.

The shopping mall web page authoring apparatus 200 inputs a plurality of images to the image classification engine 100 for each predetermined primary class, so that the image classification engine 100 classifies the images based on the primary class. Training the engine 100, and using the trained image classification engine 100, product related images to be inserted into the web page based on classifying the product-related images to be inserted into the web page for product sales according to the primary class Images can be sorted automatically.

FIG. 2 is a block diagram illustrating a configuration of the shopping mall webpage authoring apparatus 200 illustrated in FIG. 1, and FIG. 3 is a flowchart illustrating an operation flow of the shopping mall webpage authoring apparatus 200 illustrated in FIG. 2. As an example, a shopping mall web page authoring method according to a preferred embodiment of the present invention is shown.

As illustrated in FIG. 2, the shopping mall webpage authoring apparatus 200 according to an exemplary embodiment of the present invention includes an engine education unit 210, an input unit 220, a primary classifying unit 230, and a secondary classifying unit 240. ), The insertion unit 250, and the web page authoring unit 260 may be included.

1 to 3, the engine education unit 210 inputs a plurality of images to the image classification engine 100 to train the image classification engine 100 to classify each input image into a corresponding primary class. (Step: S1).

In the preferred embodiment of the present invention, the goods will be described by way of example clothing. In this case, the plurality of images may be images related to clothing items collected from a plurality of shopping malls related to clothing sales. However, this is only an example of an implementation and is not meant to be limiting. Of course, the product may be applied in various various kinds.

FIG. 4 is an exemplary diagram illustrating classifications of primary classes according to a preferred embodiment of the present invention, and FIGS. 5 to 22 are exemplary diagrams illustrating images for each primary class.

As shown in FIG. 4, in the present invention, all the images that can be used in the clothing shopping mall are suitable for the education of the image classification engine 100 and can be effectively trained without generating errors due to duplication. The tea class was derived. Such a primary class may include first to nineteenth classes.

For example, the primary first class is the whole body (body_body), the primary second class is the faceless body cut (body_only), the primary third class is the upper body (body_only), the primary fourth class is the face cut ( face_shot), 1st class 5 wears upper (wear upper_wear), 1st class 6 wears lower (wearer_wear), 1st class 7 wears shoes (wear_wear), 1st class 8 wears other Cut (etc_wear), 1st 9th class product, 1st 10th class detail, 1st 11th class spec, 1st 12th class multi , Primary 13th class is cody, primary 14th class is product_banner, primary 15th class is model_multi, primary 16th class is product_multi, 1 The seventeenth class may be other multi (etc_multi), the first 18th class may be mannequin codley, and the primary 19th class may be trash.

As illustrated in FIG. 5, the first body, the whole body, may be a product cut image in which all faces, arms, legs, hands, and feet of the model are displayed. As shown in FIG. 6, the body-only faceless cut body (body_only), which is the primary second class, may mean a product cut image in which all of the arms, legs, hands, and feet except the face are displayed on the whole body of the model. As shown in FIG. 7, the first upper class model upper_body may refer to a product cut image in which at least a portion of a face, arms, hands, and legs of the model are displayed and no feet are displayed. As illustrated in FIG. 8, the first cut class face_shot may refer to a product cut image in which the face of the model is centered and the chest is displayed.

In addition, as shown in FIG. 9, the upper fifth wear upper (upper_wear) is an image in which the chest of the model is centered so that the face and the foot are not displayed but the neck is exposed and the lower body is partially exposed. Can mean. Lower wear, which is the first sixth class (lower_wear), as shown in FIG. 10, may mean an image in which a leg or a lower body is centered, some upper body is exposed, and a foot is essentially exposed.

As illustrated in FIG. 11, the first seventh class shoe_wear may refer to an image in which only the lower body is exposed and the upper body is not exposed with the foot at the center. As illustrated in FIG. 12, the other eighth wear (etc_wear), which is the first eighth class, may refer to all product wearing images other than the first to seventh primary classes.

As illustrated in FIG. 13, the first ninth class product may refer to an image in which the overall shape of a product that is not worn is completely exposed. As illustrated in FIG. 14, the detail of the first tenth class may refer to an image in which the overall shape of the unweared product is partially exposed.

As illustrated in FIG. 15, a spec, a first eleventh class, may refer to an image made of text such as a product specification and a precaution for use. As illustrated in FIG. 16, the first twelfth class, multi, may mean an image in which two or more items having different options (eg, colors, shapes, etc.) of the same product are exposed among the product cut images. have.

As illustrated in FIG. 17, the cody, which is the first 13th class, may represent an image in which at least two different kinds of products are exposed among the product cut images and matched with other products. As illustrated in FIG. 18, a product banner (product_banner), which is a primary 14th class, may mean an image in which an entire product photograph and advertisement text are mixed.

As shown in FIG. 19, the model 15 (model_multi), which is the first 15th class, may mean an image including two or more models wearing models only. As shown in FIG. 20, the product multi (product_multi), which is the primary 16th class, may mean an edited cut image including two or more product photos or several products.

As shown in FIG. 21, the first seventeenth class refers to an image in which other multi-cuts or goods and models are mixed or edited that are not included in the aforementioned multi, model multi, and product multi categories. can do. As illustrated in FIG. 22, a mannequin cody, which is a primary 18th class, may mean a mannequin wearing image.

Meanwhile, trash, which is a primary 19th class, may mean an image that is not ambiguous or cannot be used for classification as all images outside the categories of the first to 18th classes.

The engine education unit 210 classifies the input image according to the first class and inputs it to the image classification engine 100, and the image classification engine 100 is trained based on identifying the input image.

In order to input images into the image classification engine 100, the engine training unit 210 may display an image input user interface. 23 is an exemplary diagram illustrating an image input user interface for inputting an image.

As shown in FIG. 23, the image input user interface I1 may include a display unit D1 for displaying the received images, and a selection unit T1 for selecting one primary class among a plurality of primary classes. Can be.

The image input user interface I1 may select an image corresponding to the primary class selected by the selection unit T1 among the images displayed on the display unit D1, and select the image selected by the display unit D1. The image may be input to the image classification engine 100 in correspondence with the primary class selected by T1).

For example, when the image input user interface T1 selects the other wear cut of the eighth class from the selector T1 and selects and stores at least one image corresponding to the other wear cut among the images displayed in D1, The image classification engine 100 may receive images corresponding to the eighth class cuttery cut and acquire the corresponding image classification knowledge.

As such, the engine education unit 210 may input a plurality of images to the image classification engine 100 for each primary class based on the image input user interface I1.

As mentioned above, the images input to the image classification engine 100 may be clothing-related images collected from the plurality of shopping malls SM1,..., SMn. To this end, the engine education unit 210 may collect images from product sales web pages of the plurality of linked shopping malls SM1,..., SMn.

Meanwhile, the engine education unit 210 may input a plurality of images into the image classification engine 100 for a predetermined period of time, and then calculate the number of images input for each shopping mall and the number of images input for each primary class. The engine education unit may display the calculated number of images for each shopping mall and the number of images for each primary class through a statistical user interface.

24 is an exemplary diagram illustrating a statistical user interface indicating the number of input images for each shopping mall and the number of input images for each primary class.

As illustrated in FIG. 24, the statistical user interface I2 may identify the number of images input for each shopping mall, the number of images input for each primary class, and the number of images input for each primary class among images collected from each shopping mall. Can be displayed using a graph.

In the example shown in FIG. 24, a lot of images of a specific shopping mall are relatively input, and thus the images inputted for each shopping mall are not equal, and in the case of the specific shopping mall, a relatively large number of specific primary classes are input so that images are not evenly input. I can figure it out.

The engine education unit 210 may perform image input until the number of input images for each shopping mall is equal within a preset error range based on the statistical user interface T1. In addition, the engine education unit may perform image input until the number of images input for each primary class is equal within a preset error range.

Meanwhile, the engine education unit 210 may set the minimum number of images to be input for each shopping mall or for each primary class, and may input images until the minimum number is satisfied.

When the training of the image classification engine 100 is completed, in order to produce a web page to sell products, the input unit 220 may input a product-related image to be inserted into the web page (step: S2).

When the product-related image is input, the primary classification unit 230 may automatically classify the input product-related image using the trained image classification engine 100 (step: S3). For example, the primary classifying unit 230 transmits a product related image to the image classification engine 100 to request classification of the primary class, and classifies the product related image classified by the primary class in the image classification engine 100. Can be received from.

In this way, the secondary classifying unit 240 may receive the product-related images classified for each primary class and then classify the secondary classes using the corresponding table (step: S4). In this case, the secondary class may be a group of product-related images according to positions arranged on a web page.

Each secondary class may correspond to at least one primary class. For example, assuming that a webpage is configured such that groups of images are arranged in three regions of the webpage, for example, the first region, the second region, and the third region, respectively, the image group corresponding to the first region (i.e., , The secondary first class corresponding to the primary first class, and the secondary second class corresponding to the second region correspond to the primary third class and the primary first. The secondary third class corresponding to the fifth class and the image group corresponding to the third region may be an expression corresponding to the primary second class, the primary fourth class, and the primary sixth to eighteenth classes.

The correspondence relationship between the primary class and the secondary class may be determined in a correspondence table previously stored in association with a web page. To this end, the secondary classifier 240 may set a corresponding table according to a web page, and store and manage the set corresponding table in storage.

Subsequently, the insertion unit 250 may insert product-related images classified for each secondary class into corresponding positions of the web page (step S5). That is, the image related to the second class product is inserted to be displayed in the corresponding area of the web page.

For example, in the above-described example, product-related images of the second first class are inserted into a first area corresponding to the second first class, and product-related images of the second second class are made of the second second class. The product-related images of the second third class may be inserted into the second area, and may be inserted into the third area corresponding to the second third class.

Meanwhile, the webpage authoring unit 260 may display a single screen on a display unit displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and an authoring tool capable of authoring the webpage. You can display the web page authoring screen displayed on.

25 and 26 are diagrams respectively illustrating web page authoring screens displayed by the web page authoring unit 260. FIG. 25 shows a state in which product-related images for insertion into a web page are input. The product related images are classified.

As illustrated in FIG. 25, when product-related images are input by the input unit 220, a product-related image input to a portion of the webpage authoring screen I3, for example, the display unit D2 on the left side, is previewed. It may be represented as.

The display unit D2 may variously display the input product-related images not only in the preview form but also in a small icon, a large icon, and a detailed list according to a user setting.

One side of the web page authoring screen I3 may be provided with an instruction icon C3 for instructing to automatically classify the input product-related images. When the instruction icon is selected by the user, the webpage authoring unit 260 controls the primary classifying unit 230, the secondary classifying unit 240, and the inserting unit 250 to display the inputted product-related images 1. It can classify by class, classify it again by class and insert it into the corresponding position of web page.

As illustrated in FIG. 26, when the instruction icon C3 is selected, the first region D3, the second region D4, and the third region D5 of the webpage authoring screen I4 are classified by secondary classes. The inserted product-related images are displayed in a preview form. The classified product-related images may be displayed in various ways, such as a small icon, a large icon, and a detailed list.

Although not shown, a text input unit may be provided at an adjacent portion of the classified area of FIG. 26 to input related text (eg, description of an inserted product related image). Therefore, while authoring a web page, you can see the image of the product that is automatically classified and inserted with just one click while viewing the form of the web page, and by entering the relevant text, you can create a web page for product sales quickly and conveniently. do.

While the present invention has been described by way of example with reference to the preferred embodiments thereof, those skilled in the art will variously modify and change the present invention without departing from the technical matters and scope of the present invention as set forth in the claims below. It will be understood that it can be done. Accordingly, modifications to future embodiments of the present invention will not depart from the technology of the present invention.

1: Shopping Mall Web Page Authoring System
100: image classification engine
200: shopping mall web page authoring device
210: Engine Education Department
220: input unit
230: primary classification unit
240: secondary classification unit
250: insertion part
260: Web page authoring department

Claims (23)

In the method performed by the shopping mall web page authoring apparatus,
Training an image classification engine to input a plurality of images into an image classification engine so as to classify each input image into a corresponding primary class;
Inputting product related images related to a product to be sold through a web page;
Classifying the input product-related images by primary class using the educated image classification engine;
Classifying the product-related images classified by the primary class by the secondary class based on a corresponding table; And
Inserting the product-related images classified by the second class into corresponding positions of the web page;
The primary class is
The first class, which is a product cut image in which the face, arms, legs, hands, and feet of the model are all displayed, and the first class, which is a product cut image in which all the arms, legs, hands, and feet except the face are displayed on the model body. The first third class, a product cut image with at least a portion of the face, arms, hands, and legs displayed and no feet, and the first fourth class, a product cut image with the face of the model centered and the chest displayed. The 5th class, an image where the chest is centered and the face and feet are not visible but the neck is exposed and the lower part of the body is partially exposed. Goods other than the first sixth class, the first seventh class, the first seventh class, the first to seventh class, the image is only the lower half of the body and the upper body is not exposed with the feet centered The first eighth class, the wearing image, the first ninth class, the image of which the overall shape of the unweared product is fully exposed, the first tenth class, the image of the entire form of the unweared product, partially exposed, the product-related text The first 11th class, which is an image consisting of two or more products of different product options, each of which has different options for the same product. The first 13th class, which is an image that matches another product as an image, the 1st 14th class, which is an image of the whole product and the ad text, and the 1st 15th class, which is an image containing two or more models wearing models. The first 16th class, the 1st 9th class, the 1st 15th class, and the 1st 16th class, the photograph being an edited cut image containing two or more or several products Mall Web page authoring method comprising the multi-cut that does not include a category or product and model mix or edit the image in at least one of the first 17 classes and a mannequin wearing the image of the first 18 classes.
The method of claim 1, wherein training the image classification engine comprises:
A display unit for displaying the received images and a selection unit for selecting one primary class among a plurality of primary classes, and designating an image corresponding to the primary class selected by the selection unit among the images displayed on the display unit Displaying a user interface inputable to the image classification engine; And
And inputting the plurality of images into the image classification engine for each primary class based on the user interface.
The method of claim 1, wherein the plurality of images are clothing-related images collected from a plurality of shopping malls.
Training the image classification engine,
And inputting the clothing goods-related images into the image classification engine until the clothing goods-related images for each shopping mall are equalized within a set number of ranges.
The display apparatus of claim 3, further comprising: displaying a user interface indicating at least one of information indicating how many clothing goods-related images have been input to an image classification engine for each shopping mall, and information for comparing the number of images of clothing goods for each shopping mall; Mall Web page authoring method further comprising the step.
The method of claim 1, further comprising setting at least one primary class corresponding to each secondary class,
Wherein each of the secondary classes is a group of product-related images arranged at a predetermined position on the web page.
The web page of claim 1, further comprising: a display unit for displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and a web authoring tool for authoring the webpage. Further comprising displaying an authoring screen,
And when the indication icon is selected by a user, classifying by the primary class, classifying by the secondary class, and inserting the shopping mall.
7. The method of claim 6, wherein the web page authoring screen displays the product-related images inserted into the second class in the authoring tool in the form of a preview in a corresponding arrangement position in real time.
An engine education unit for training an image classification engine to input a plurality of images into an image classification engine and classify each input image into a corresponding primary class;
An input unit for inputting product related images related to a product to be sold through a web page;
A primary classifying unit which classifies the input image related products by primary classes using the educated image classification engine;
A secondary classification unit classifying the product-related images classified by the primary class by the secondary class based on a corresponding table; And
An insertion unit for inserting the product-related images classified by the second class into corresponding positions on the authoring screen of the web page,
The primary class is
The first class, which is a product cut image in which the face, arms, legs, hands, and feet of the model are all displayed, and the first class, which is a product cut image in which all the arms, legs, hands, and feet except the face are displayed on the model body. The first third class, a product cut image with at least a portion of the face, arms, hands, and legs displayed and no feet, and the first fourth class, a product cut image with the face of the model centered and the chest displayed. The 5th class, an image where the chest is centered and the face and feet are not visible but the neck is exposed and the lower part of the body is partially exposed. Goods other than the first sixth class, the first seventh class, the first seventh class, the first to seventh class, the image is only the lower half of the body and the upper body is not exposed with the feet centered The first eighth class, the wearing image, the first ninth class, the image of which the overall shape of the unweared product is fully exposed, the first tenth class, the image of the entire form of the unweared product, partially exposed, the product-related text The first 11th class, which is an image consisting of two or more products of different product options, each of which has different options for the same product. The first 13th class, which is an image that matches another product as an image, the 1st 14th class, which is an image of the whole product and the ad text, and the 1st 15th class, which is an image containing two or more models wearing models. The first 16th class, the 1st 9th class, the 1st 15th class, and the 1st 16th class, the photograph being an edited cut image containing two or more or several products Comprising a multi-cut that does not include a category or product and at least one of the model mix or edit the image in the first 17 classes and a mannequin wearing the image of the first 18th class shopping mall web page authoring apparatus.
The method of claim 8, wherein the engine training unit,
A display unit for displaying the received images, and a selection unit for selecting one primary class among a plurality of primary classes, and designating an image corresponding to the primary class selected by the selection unit among the images displayed on the display unit; Displays a user interface that can be entered into the image classification engine,
The shopping mall web page authoring apparatus, characterized in that for inputting the plurality of images into the image classification engine for each primary class based on the user interface.
The method of claim 8, wherein the plurality of images is a clothing product-related images collected from a plurality of shopping malls
The engine training unit,
The shopping mall web page authoring apparatus, wherein the clothing goods-related images are input to the image classification engine until the shopping goods-related images are equalized within the set number.
The method of claim 10, wherein the engine training unit,
A shopping mall web displaying a user interface representing at least one of information indicating how many clothing goods-related images have been input to an image classification engine and information for comparing the number of clothing goods-related images for each shopping mall Page authoring device.
The method of claim 8, wherein the secondary classifier sets at least one primary class corresponding to each secondary class,
Each second class is a shopping mall webpage authoring apparatus, characterized in that a group of product-related images arranged at a predetermined position on the webpage.
The web display device of claim 8, further comprising: a display unit for displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and a web for displaying the authorable authoring tool of the webpage on a single screen. Further comprising a web page authoring unit for displaying the page authoring screen,
When the instruction icon is selected by the user, the web page authoring unit controls the primary classifying unit, the secondary classifying unit, and the inserting unit to classify the input product-related images by the primary class, and The shopping mall web page authoring apparatus, characterized in that classified by car class and automatically performing a process of inserting the corresponding position of the web page.
The shopping mall web of claim 13, wherein the webpage authoring unit displays, in real time, a preview form of product-related images inserted for each secondary class in an authoring tool of an authoring screen of the webpage in a corresponding arrangement position. Page authoring device.
Image classification engine; And
Train the image classification engine to classify each input image into the corresponding primary class by inputting a plurality of images into the image classification engine, input product related images related to the product to be sold through the webpage, and And classify the input product-related images by primary class, classify the product-related images classified by the primary class by secondary class based on a corresponding table, and use the image classification engine. A web page authoring apparatus for inserting the product-related images classified by car class into corresponding positions in the authoring screen of the web page,
The primary class is a primary cut that is a product cut image in which all of the face, arms, legs, hands, and feet of the model are displayed. Class 2, a primary cut, a product cut image in which at least a portion of the model's face, arms, hands, and legs are displayed and no feet appear. Class 4, model's chest is centered, face and feet are not visible but neck is exposed and lower body is part 5, first class, legs or lower body is centered, some upper body is exposed and feet are essential The first sixth class, which is an image exposed to the first sixth class, the first seventh class, the first to seventh primary The first 8th class, which is an image of wearing a product other than Lath, the first 9th class, which is an image in which the overall shape of a product that is not worn is fully exposed, and the first 10th class, which is an image in which the overall shape of a product that is not worn is partially exposed. , The 11th class, which is an image of product-related text, and the 2nd class of the 12th class, which is an image in which two or more products with different options of the same product are exposed. The first 13th class, which is an image that is exposed more than and matched with other products, the 1st 14th class, which is a mixture of the whole product photograph and the advertising text, and the 1st first image, which contains two or more models wearing models. 15th class, the first 16th class, the first 9th class, the 1st 15th class, prize Multi-cut or goods and models not included in the category of the primary 16 class includes at least one of the primary 17 class, which is a mixed or edited image and the primary 18 class, which is a mannequin wearing image Mall webpage authoring system.
delete delete delete delete delete delete An engine education unit for training an image classification engine to input a plurality of images into an image classification engine and classify each input image into a corresponding primary class;
An input unit for inputting product related images related to a product to be sold through a web page; And
Displaying a display unit for displaying the inputted product-related images, an instruction icon for instructing to automatically classify the inputted product-related images, and a webpage authoring screen for displaying the authoring tool of the webpage on a single screen; When the indication icon is selected by the user, the input product-related images are classified by the primary class, the product-related images classified by the primary class are classified by the set secondary class, and the secondary class. It includes a web page authoring unit for automatically performing the process of inserting the product-related images classified by each in the corresponding position of the web page,
The primary class is a primary cut that is a product cut image in which all of the face, arms, legs, hands, and feet of the model are displayed. Class 2, a primary cut, a product cut image in which at least a portion of the model's face, arms, hands, and legs are displayed and no feet appear. Class 4, model's chest is centered, face and feet are not visible but neck is exposed and lower body is part 5, first class, legs or lower body is centered, some upper body is exposed and feet are essential The first sixth class, which is an image exposed to the first sixth class, the first seventh class, the first to seventh primary The first 8th class, which is an image of wearing a product other than Lath, the first 9th class, which is an image in which the overall shape of a product that is not worn is fully exposed, and the first 10th class, which is an image in which the overall shape of a product that is not worn is partially exposed. , The 11th class, which is an image of product-related text, and the 2nd class of the 12th class, which is an image in which two or more products with different options of the same product are exposed. The first 13th class, which is an image that is exposed more than and matched with other products, the 1st 14th class, which is a mixture of the whole product photograph and the advertising text, and the 1st first image, which contains two or more models wearing models. 15th class, the first 16th class, the first 9th class, the first 15th class, prize Multi-cut or goods and models not included in the category of the primary 16 class includes at least one of the primary 17 class, which is a mixed or edited image and the primary 18 class, which is a mannequin wearing image Shopping mall web page authoring device. delete

Images