KR100933269B1 - Search method and system using color keyword - Google Patents

Abstract

The present invention relates to a search method and system using a color keyword, the method comprising the steps of extracting a color code corresponding to the color keyword with reference to the color table, a database of images or videos having the extracted color code as a representative color code Generating a first search result by extracting from the image, and extracting an image or video having a general keyword as meta information from a database to generate a second search result. According to the present invention, an image or a video desired by a user can be extracted and provided to the user.

Color keyword, color code, representative color code, database, search server

Description

Search method and system using color keyword {METHOD AND SYSTEM FOR SEARCHING USING COLOR KEYWORD}

The present invention relates to a search method and system, and more particularly, to a search method and system using color keywords.

With the rapid spread of high speed internet in recent years, it has become an indispensable necessity in modern life. Most users who use the Internet access the Internet through a browser and search for information through an Internet portal site. When the user enters a search word in the search box of the Internet portal, the Internet portal extracts various information corresponding to the search word from the database and provides the search result to the user. The extracted information is divided into categories such as dictionaries, knowledge information (eg, Naver's knowledge iN), blogs, cafes, specialty materials, sites, books, news, web pages, images, videos, and the like, and is delivered to the user for display. When the user selects any of the displayed information, the Internet portal extracts the document from the database and displays it to the user.

In general documents, the search technology has been advanced to extract the search results that users want, but it is difficult to extract the search results that users want. In other words, when searching for an image or video, the search is performed based on the characters in the title or description of the image or video. Therefore, if the description of the image or video is not sufficiently described, the user cannot obtain the desired search result. .

Accordingly, an aspect of the present invention is to provide various methods for extracting a representative color code value from an image or a video.

In addition, another technical problem to be achieved by the present invention is to provide a search method and system that can extract the image or video desired by the user.

A search method according to an aspect of the present invention for achieving the technical problem is a search method for a search server to perform a search for a search query including a color keyword and a general keyword, corresponding to the color keyword with reference to a color table Extracting a color code to generate a first search result by extracting an image or a video having the extracted color code as a representative color code from a database, and generating an image or video having the general keyword as meta information; Extracting from a database to generate a second search result.

The image or video of the second search result may be an image or video having the general keyword as meta information among the image or video of the first search result.

The image or video of the first search result may be an image or video having the extracted color code as a representative color code among the images or video of the second search result.

Extracting an image or a video having the color keyword and the general keyword as meta information from the database to generate a third search result, and generating an image or video of the first and second search results and the third search result. The method may further include removing a duplicate image or a video from the image or the video.

The method may further include calculating a share of the image or video for each color code included in the image or the video, and determining the representative color code for the image or the video based on the share.

Grouping color codes corresponding to the same color keyword among the color codes included in the image or the video by referring to the color table, and calculating a share of the image or the video for each of the bundles based on the share; The method may further include determining a representative color code for the image or video.

A computer readable medium according to another aspect of the present invention records a program for causing a computer to execute any of the above methods.

A search system according to another aspect of the present invention, receives a search query including a database, a color keyword and a general keyword that stores an image or video and stores meta information and representative color codes of the image or video, and the color keyword. A search server for extracting a color code corresponding to, extracting an image or video having the extracted color code as a representative color code from the database, and extracting an image or video having the general keyword as meta information from the database; Include.

As described above, according to the present invention, a color code may be extracted from an image and a video, and a representative color code of the image and a video may be extracted in various ways based on the share of the color code.

In addition, according to the present invention, by extracting the representative color code value from the image and video, and extracts the image or video having a color code corresponding to the color keyword input by the user as the representative color code can provide a user desired search results have.

DETAILED DESCRIPTION Embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention.

First, a search system according to an exemplary embodiment of the present invention will be described in detail with reference to FIGS. 1 and 2.

1 is a block diagram illustrating a search system and a user terminal connected thereto according to an embodiment of the present invention, and FIG. 2 is a diagram illustrating an example of a color table of the color information database illustrated in FIG. 1.

As shown in FIG. 1, the search system 300 according to an embodiment of the present invention includes a search server 310, a color information database 320 connected to the search server 310, an image database 330, and a video. The database 340 includes a color code extraction server 350 connected to the image database 300 and the video database 340, and is connected to the plurality of user terminals 100 through the communication network 200.

The user terminal 100 transmits and receives information by accessing the search server 310 through the communication network 200. The user terminal 100 includes a microprocessor as well as a desktop computer and includes memory means such as a laptop computer, a workstation, a palmtop computer, a personal digital assistant (PDA), a web pad, a mobile communication terminal, and the like. It can be made of a terminal equipped with a computing power.

The user terminal 100 may include a communication module (not shown) and an application program for accessing the search system 300 through the communication network 200. The application program accesses the search system 300 through the communication network 200 according to a user's operation and performs various functions. An example of the application may include, but is not limited to, a web browser such as Microsoft's Internet Explorer.

The communication network 200 may include a telephone network as well as a data communication network including a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), the Internet, and the like. It can be used, and it does not matter wired and wireless, and it does not matter which communication method is used.

The search server 310 receives a search query input by the user from the user terminal 100, parses the search query, and extracts a color keyword and a general keyword therefrom. The color information database 320, the image database 330, and the video database 340 are searched using the color keyword and the general keyword, and the search result is provided to the user terminal 100.

The search query entered by the user includes a color keyword and a general keyword. Color keywords are words related to color, for example, nouns such as 'red' and 'red', and adjectives such as 'red (red)' and 'red (red)'. General keywords are words to be searched. An example of a search query that combines a color keyword and a general keyword is 'green field'. The search query may include a plurality of color keywords, such as 'red roses and yellow sunflowers.' Even when the search word is a sentence, if the sentence includes color information, the search server 310 extracts the color keyword. Search can be performed.

The color information database 320 stores the color table. The color table includes a plurality of color keywords and color codes corresponding to each color keyword. The color keyword may correspond to at least one color code, and synonyms may be processed between the color keywords.

The color code is used to represent the color of the pixels forming each frame of the image and the video as digital data, and may be a combination of primary colors. Examples of the primary colors include three primary colors such as red (R), green (G), and blue (B), and the color codes of each primary color are 8 bit data values, that is, 256 values between 00 and FF (hex). It can have any one of the values. However, the color code may use a primary color consisting of colors other than red, green, and blue, for example cyan (C), magenta (M), and yellow (M). May have

For example, referring to the color table of FIG. 2, the color keywords CK1, which may be represented by red, red, red, red, and the like, are red (R) color codes D0 to FF (hex) and green (G) colors. Corresponds to codes 00 to 2F (hex) and blue (B) color codes 00 to 2F (hex). The color keyword (CK2) that can be expressed as a green color is red (R) color code 00 to 2F (hex), green (G) color code D0 to FF (hex), and blue (B) color code 00 to 2F. corresponds to (hex). The color keyword CK3 that can be expressed as a blue color is red (R) color code 00 to 2F (hex), green (G) color code 00 to 2F (hex), and blue (B) color code D0 to FF. corresponds to (hex). Of course, a color code corresponding to each of the color keywords CK1, CK2, and CK3 may be set differently.

The image database 330 stores a plurality of images and image information associated with each image. The image information includes a title of the image, an image category, a tag, an image description, and the like. The image information is stored as meta information that the search server 310 can easily query the image database 330. The image database 330 also stores a representative color code for each image.

The video database 340 stores a plurality of videos and video information associated with each video. The video information includes a video title, a video category, a tag, a video description, a caption, and the like. The video information is stored as meta information that the search server 310 can easily search the video database 330. The video database 340 also stores a representative color code for each video. Unlike an image, a video is composed of many frames, and thus, a video may have a representative color code for each of a plurality of frames.

When a user inputs an image or a video and related information through the user terminal 100, the image or video processing server (not shown) may store it in the image database 330 or the video database 340. In addition, a search robot (not shown) may search for an image or video that is publicly available on the Internet, and store information related to the image or video in the image database 330 or the video database 340.

The color code extraction server 350 extracts a representative color code for each image stored in the image database 330 and stores the extracted representative color code in the image database 330 in association with the meta information of the corresponding image. In addition, the color code extraction server 350 extracts the representative color code for the video stored in the video database 340 and stores the extracted representative color code in association with the meta information of the corresponding video in the video database 340. . Here, the representative color code means a color code of the dominant color in the image and video.

The color code extraction server 350 may extract the representative color code from the image and the video based on various criteria as follows. For convenience of explanation, a method of extracting a representative color code from an image will be described first.

The color code extraction server 350 receives one image from the image database 330 and extracts a color code of each pixel constituting the image. No matter what format the image is stored in, the color code extraction server 350 may extract the color code from each pixel of the image in a suitable manner.

The color code extraction server 350 may calculate a share occupied in the image for each color code and determine a representative color code based on the share. The share of one color code is the number of times that color code appears in the image divided by the total number of pixels.

As a first criterion, the color code extraction server 350 may determine the color code having the highest occupancy as the representative color code. In this case, there is one representative color code. As a second criterion, the color code extraction server 350 may set a predetermined number of color codes as a representative color code, starting with the color code having the highest occupancy rate. Then, a plurality of representative color codes may be extracted instead of one representative color code. As a third criterion, the color code extraction server 350 may determine a color code having a share of a predetermined ratio or more as a representative color code. In this case, if the occupancy rate of the color code having the highest occupancy is smaller than the predetermined ratio, the corresponding image has no representative color code. As a fourth criterion, the color code extraction server 350 may set a color code that is an intersection of the second criterion and the third criterion as the representative color code.

On the other hand, the color code extraction server 350 refers to the color table of the color information database 320 and bundles the color codes corresponding to the same color keywords among the color codes shown in the image, and represents the representative colors based on the share of each bundle. You can decide on the code. Here, the bundle occupancy is the number of times the color code in the bundle appears in the image divided by the total number of pixels.

The color code extraction server 350 applies the first to fourth criteria described above for each bundle to determine a bundle suitable for the criteria. That is, a bundle having the highest occupancy rate is determined, a predetermined number of bundles are determined in order from a bundle having the highest occupancy rate, a bundle having a share having a predetermined ratio or more, or a share having a predetermined ratio or more, but having the highest share. Branches can define a predetermined number of bundles in order from the bundle.

Then, the color code having the largest occupancy in the bundle thus obtained can be determined as the representative color code for the image.

The color code extraction server 350 may extract the representative color code according to at least one of the criteria described above as needed. In addition to these criteria, the representative color code may be determined based on various criteria.

The color code extraction server 350 may extract a representative color code for each frame of the video or extract a representative color code from a key frame of the video. In general, a video is compressed and stored on the basis of a key frame. For the key frame, the entire information of the frame screen is stored, but only the information of a part different from the corresponding key frame is stored for the frames up to the next key frame. The key frame may be designated every time (for example, 5 seconds or 10 seconds) or according to a change in the screen, that is, frequently specified in a portion where the change of the motion is severe and less in a portion where the change of the motion is small. Therefore, extracting the representative color code from the key frame can improve the data processing speed for extracting the representative color code, and when the key frame is specified based on the screen change, extraction of the representative color code for the overlapping screen is relatively It is possible to extract a representative color code of a screen that is mainly representative of a video.

Since the method of extracting the representative color code from the frame of the video is substantially the same as the method of extracting the representative color code from one image, a detailed description thereof will be omitted.

The color code extraction server 350 may extract the representative color code as soon as the user registers the image or video. In addition, the color code extraction server 350 may extract the representative color code in a batch processing method. That is, as time passes, the image or video registered by the user is accumulated in the image database 330 or the video database 340. When the new image or video is accumulated in the database 330 or 340 by a certain amount, the newly accumulated image is recorded. Alternatively, the representative color code can be extracted from the video. Alternatively, the color code extraction server 350 may extract a representative color code for a newly accumulated image or video in the image database 330 or the video database 340 at a specific time period.

Next, a search method using color keywords of the search system according to an exemplary embodiment of the present invention will be described in detail with reference to FIG. 3.

3 is a flowchart illustrating a search method according to an embodiment of the present invention.

As shown in FIG. 3, when a user inputs a search query to a search page displayed on the user terminal 100 and makes a search request, the search system 300 receives the search query (S410). The search server 310 parses the received search query (S420) and determines whether a color keyword is included in the search query (S430). The search server 310 may determine whether a color keyword is included in the search query by determining whether the parsed word exists in the color keyword of the color information database 320.

The search system 300 may include a separate web server (not shown) to provide a search page to the user terminal 100, and the web server may receive a search query and transmit the search query to the search server 310. Alternatively, the web server may parse the search query to classify the color keyword and the general keyword and transmit the color keyword to the search server 310.

If it is determined in step S430 that no color keyword is included, the search system 300 returns to operation. In this case, a separate search server (not shown) that performs a general keyword search may perform a general search and provide a search result to the user terminal 100.

If it is determined in step S430 that a color keyword is included, the search server 310 extracts a color code corresponding to the color keyword from the color information database 320 (S440).

The search server 310 extracts an image or video having the color code extracted in step S440 as the representative color code from the image database 330 or the video database 340 (S450). Then, the search server 310 searches for image information or video information about the image or video extracted in step S450 and finds an image or video having information corresponding to the general keyword of the search query among the images or videos. (S460).

Steps S450 and S460 may be performed in a different order. That is, after performing a search for a general keyword, an image or video may be extracted as a search result, and an image or video having a color code corresponding to the color keyword as a representative color code may be extracted from the extracted image or video. .

In step S450 and S460, the image database 330 or the video database 340 is searched for the entire search query without distinguishing the color keyword and the general keyword (S470). That is, an image or a video including both a color keyword and a general keyword is extracted from the image information of the image database 330 or the video information of the video database 340. By extracting these images or videos together and using them in search results, the accuracy of the search can be further improved.

After comparing the search results in steps S450 and S460 with the search results in step S470, the duplicated results are removed, and the final search result is provided to the user terminal 100 (S480).

Even when the user inputs a plurality of color keywords, the search system 300 may provide a search result in the above manner. That is, an image or video having a color code corresponding to each color keyword as a representative color code is extracted, and a search result including a general keyword is extracted therefrom. In addition, an image or a video having a color code corresponding to a plurality of color keywords as a representative color code is extracted, and a search result including a general keyword is extracted therefrom. Duplicates are removed from these search results and the final search result is provided to the user terminal 100.

When the search system 300 receives a search query such as, for example, 'red roses and yellow sunflowers', the search system 300 searches for the 'red roses' and 'yellow sunflowers' according to the search method according to the embodiment of the present invention. To perform. In addition, it extracts an image or video having a color code corresponding to the color keyword 'red' and a color code corresponding to the color keyword 'yellow' as a representative color code, and extracts the general keywords 'rose' and 'sunflower' from the extracted image or video. Extract an image or video with information that includes. Then, the duplicate search is removed from the two search results and the final search result is provided to the user terminal 100.

On the other hand, the user can perform a search using only the color keyword. In this case, an image or a video having a color code corresponding to the color keyword as a representative color code may be extracted and provided as a search result.

As described above, according to the search system 300 according to an exemplary embodiment of the present invention, a representative color code of the image and the video may be extracted from the image and the video stored in the image database 330 and the video database 340 in various ways. have. In addition, when a user wants to find an image or video including a specific color, the user can search and provide an appropriate image or video. That is, if a user wants to find an 'green field' image, for example, if the image exists in the image database 330 and the image information includes 'field', the image information does not have the keyword 'green'. Since the image having the color code corresponding to 'green' as the representative color code may be extracted from the image database 330, the 'green field' image may be provided to the user.

Embodiments of the invention include a computer readable medium containing program instructions for performing various computer-implemented operations. This medium records a program for executing the retrieval method using the retrieval option layer described above. The media may include, alone or in combination with the program instructions, data files, data structures, and the like. Examples of such media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CDs and DVDs, floppy disks and program commands such as magnetic-optical media, ROM, RAM, flash memory, and the like. Hardware devices configured to store and perform such operations. Alternatively, the medium may be a transmission medium such as an optical or metal wire, a waveguide, or the like including a carrier wave for transmitting a signal specifying a program command, a data structure, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

Although the preferred embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

1 is a block diagram illustrating a search system and a user terminal connected thereto according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of a color table of the color information database shown in FIG. 1.

3 is a flowchart illustrating a search method according to an embodiment of the present invention.

<Description of Drawing>

100: user terminal, 200: communication network,

300: search system, 310: search server,

320: color information database, 330: image database,

340: video database, 350: color code extraction server

Claims (14)

A search method in which a search server performs a search on a search query including a color keyword and a general keyword, Extracting a color code corresponding to the color keyword by referring to a color table; Generating a first search result by extracting an image or a video having the extracted color code as a representative color code from a database; and Extracting an image or a video having the general keyword as meta information from the database to generate a second search result; Search method comprising a. In claim 1, And the image or video of the second search result is an image or video having the general keyword as meta information among the images or video of the first search result. In claim 1, And the image or video of the first search result is an image or a video having the extracted color code as a representative color code among the image or video of the second search result. The method of claim 2 or 3, Generating a third search result by extracting an image or a video having the color keyword and the general keyword as meta information from the database; and Removing duplicate images or videos from the images or videos of the first and second search results and the images or videos of the third search result. Search method comprising more. In claim 1, Calculating a share occupied in the image or video for each color code included in the image or video and determining the representative color code for the image or video based on the share. In claim 1, Grouping the color codes corresponding to the same color keywords among the color codes included in the image or the video by referring to the color table, and Calculating a share of the image or video for each of the bundles, and determining a representative color code for the image or video based on the share; Search method comprising more. A computer-readable medium having recorded thereon a program for causing a computer to execute the method of any one of claims 1 to 3 and 5 to 6. A database that stores an image or video and stores meta information and representative color codes of the image or video, Receiving a search query including a color keyword and a general keyword, extracting a color code corresponding to the color keyword, extracting an image or video having the extracted color code as a representative color code from the database, and then performing a first search result A search server for generating a second search result by extracting an image or a video having the general keyword as meta information from the database; Search system comprising a. In claim 8, And the image or video of the second search result is an image or video having the general keyword as meta information among the image or video of the first search result. In claim 8, And the image or video of the first search result is an image or video having the extracted color code as a representative color code among the images or video of the second search result. The method of claim 8 or 9, The search server, Extracting an image or a video having the color keyword and the general keyword as meta information from the database to generate a third search result, and generating an image or a video of the first and second search results and an image or a video of the third search result A search system for removing duplicate images or videos from a video. In claim 8, A color code extraction server for calculating a share occupied in the image or video for each color code included in the image or video and determining the representative color code for the image or video based on the share Search system further comprising. In claim 8, Bundling the color codes corresponding to the same color keyword among the color codes included in the image or the video by referring to the color table, calculating the occupancy occupied in the image or the video for each bundle, and based on the share Or a color code extraction server that determines a representative color code for a video Search system further comprising. A computer-readable medium having recorded thereon a program for executing the method of claim 4 on a computer.

Images