KR20070014219A - Automatic web source creation system - Google Patents

Abstract

A method for automatically detecting goods from an image and the method for generating a related web source are provided to generate the web source for connecting to a webpage including information matched with the inputted goods by detecting the goods from the image of an Internet shopping mall. An image file is loaded to generate a goods information source code(s500). A goods image is extracted from the image file(s510). Link information is connected to the extracted image(s520). The link information is stored to generate a link information source for connecting to the extracted goods image(s530).

Description

How to automatically detect merchandise in an image and how to create an associated web source for a product

1 is a block diagram showing the configuration of a method for automatically detecting a product in an image of the present invention and a method and system for generating address information of a web page associated with a product.

2 is a flowchart illustrating a process of displaying an image file according to an exemplary embodiment.

3 is a flowchart illustrating a process of detecting a product region in an image file according to an embodiment of the present invention.

4 is a flowchart illustrating a process of connecting link information to an extracted product image according to an embodiment of the present invention.

5 is a flowchart illustrating a process of generating source code connecting link information to an extracted product image according to an embodiment of the present invention.

6 is a flowchart illustrating each procedure according to an embodiment of the present invention.

The present invention relates to a technology for extracting a corresponding image for each product from an image of an internet shopping mall and generating information of a web page address corresponding to each product. More specifically, when a model wears a product to sell, the model wears a variety of products in addition to the product, and automatically recognizes the products so that they can be moved to a web page corresponding to each product in the image. And a method and system for automatically generating a source that can be linked to a web page of a corresponding product.

The internet shopping mall provides an image of a product to be sold and a model to which the product is applied. In this case, the model wears other products in addition to the product to be sold, and most of the information about the product is not described or added to the information page of the product to be sold or a new image is attached to be linked. In this case, additional search and verification is required to obtain information on each product.

When a user clicks a region on an image, a technique of linking to another web page can be implemented by linking to the image through an image map, cutting the image by a specific region, and linking to the image.

There is a method of creating an image map for an image using a web editor, but it is difficult to know the link information of the corresponding product one by one. have

Accordingly, the present invention for solving the above problems is to provide a method and system that can generate a source code for connecting a web page containing information corresponding to the input product.

Other objects of the present invention will be easily understood through the description of the following embodiments.

In order to achieve the above object, the product image detection and source code generator according to the present invention is a device for processing the image data and generates a product information link source for the image data

First procedure for importing an image file for generating product information source code

Second procedure of the processing method for extracting the product image in the image file

Third procedure for linking link information to an extracted image

A fourth procedure for storing the link information source associated with the extracted image.

It is characterized by including.

Before describing the embodiments, an overview of the present invention will be described.

According to the present invention, the product image is automatically extracted and the source code is automatically generated to allow the application to conveniently create and deliver a product information link to both the consumer and the seller.

(Example)

Hereinafter, with reference to the drawings will be described embodiments of the present invention;

1 is a block diagram showing the configuration of an application operation according to an embodiment of the present invention. The product image detection and source code generator includes the host computer (100), the CPU (101), the HDD (102), the application software (103), the I / O (104), the monitor (105), the keyboard (106), and the mouse (107). ), Consisting of a DB server 108.

In particular, the application software 103 is software, such as a word processor, spreadsheet, or internet browser of the host computer 100, and is also the present invention. I / O is responsible for the input and output of the monitor, keyboard, and mouse. The HDD 102 stores various types of image files.

(First procedure)

2 is a flow chart showing the flow of the first procedure. The flowchart is executed by the CPU 101 of the host computer 100.

In step S101, an image file in the HDD 102 is selected. In step s102, it is checked whether the image file format is a standardized image file, and if it is a standard image file, the loaded image is displayed on the monitor 105.

(Second procedure)

3 is a flow chart showing the flow of the second procedure. The flowchart is executed by the CPU 101 of the host computer 100.

Product image extraction is performed on the image file selected in step S210, that is, the file called in the first procedure. Extraction methods include extracting product images through RGB values, extracting product images through HSI values, extracting product images by detecting edges, and extracting product images through pattern searches. In the drawing, a method (s220) of extracting a product image based on an RGB value is taken as an example. In the product image extracted by the RGB value, the area of the image extracted in step S230 is displayed on the monitor 105 as a dotted line.

(Third procedure)

4 is a flow chart showing the flow of the third procedure. The flowchart is executed by the CPU 101 of the host computer 100.

In step s310 of the third procedure, link information is input to an automatically extracted product image area s300 which is a result of performing the first procedure and the second procedure, and information corresponding to each extracted product image is linked. Let's do it.

(4th procedure)

5 is a flow chart showing the flow of the fourth procedure. The flowchart is executed by the CPU 101 of the host computer 100.

The fourth procedure links the link information to the extracted product image which is a result of performing the first procedure, the second procedure, and the third procedures in step S400.

If the user selects save, the fourth procedure is completed after generating the source code connecting the link information to the extracted product image.

If not saved, execution of the fourth procedure is terminated immediately.

(Execution of the first, second, third, and fourth procedures)

Hereinafter, the execution of the first, second, third, and fourth procedures in the case where the image processing method is an application that can be operated in, for example, Microsoft's Windows XP operating system, will be described.

When the image processing is executed directly, the first procedure will be approved first, and then the second, third, and fourth procedures will be approved.

When an image file is selected from a storage device such as the hard disk 101, the image is displayed on the monitor 105 and the image processing method is instructed when the second procedure is started. The third procedure is approved and the information connected to the extracted image. Will be entered. When the fourth procedure is started, link information is linked to the extracted product image. When storage is ordered, it generates a source code linking the web page corresponding to the entered product.

In an embodiment of the present invention, the executable file for the application implementing the image processing method is named "WaColorProcess.exe".

When a start command is received by double-clicking "WaColorProcess.exe", the first procedure execution screen is first displayed on the monitor 105 because the first to fourth procedures should be executed.

If an image file is selected from a storage device such as a hard disk in step S100, it is determined whether the image file is in step S110. If the image file is an image file, the image is displayed on the monitor in step S120.

If a color image of the submenu is clicked, a second procedure is started and the selected image file s200 is extracted at step s210 with the product image based on the RGB value. In operation S220, the extracted image area is displayed on the monitor.

When the third procedure starts, the extracted product image area is clicked and the linked information is input in step S310.

If the link information of the image is linked (s400) and stored in step s410, the source code is generated and ends.

(Other embodiments)

As an example of the above-described embodiment of the present invention, the case of extracting the product image region extraction method through the RGB value has been described. However, the present invention is not limited to the method of extracting the RGB image through the RGB value. Various methods can be used, such as a method of extracting the protruding point of the boundary region of the product image and the background, a method through a pattern search.

In the above example, when the DB server 108 is not established, but the DB server 108 is constructed, the procedure of Procedure 1 and Procedure 2 is the same as the above example, and the link in step S310 of Procedure 3 is performed. Enter key value (tag information) saved in DB, not information input. Product information is connected to the DB server, not the link information connection of the product image extracted in step (s400) of Procedure 4.

The present invention automatically detects a region for a product in a product image and automatically generates a web source for each product, thereby dividing the region for each product through a web editor and remembering a link address. Eliminates the process that needs to be modified when changing or deleting a. This enables intuitive and immediate navigation to the information page for each product within a single image, minimizing the additional process of searching for information pages for each product.

Claims (3)

In the second procedure, Each product image area in the registered image is automatically extracted by using RGB information, HCI information, edge information, wavelet transform method, and the like. How is the screen displayed on the screen? In the third procedure, Selecting each product image area within the displayed image, linking to and storing information pages corresponding to each selected product image In the above connection method, How to register key related to each product when database for each selected product is established How to register the address of the information page corresponding to each product if there is no database for each selected product. In the fourth procedure, How to create a web source that can link to the information page corresponding to each product by using the link address and related key value saved as a result of the third procedure

Images