KR101952622B1 - Method, apparatus, system and computer program for image retrieval based on local color difference - Google Patents

Abstract

The invention relates to a method of processing an image in a service server, the method comprising: collecting a plurality of images and extracting a feature region of the collected image; Extracting pixels of the feature region by skipping at predetermined intervals; Generating data on a change value of a color of adjacent pixels as a descriptor; And merging the descriptors to generate index values, and indexing the index values.

Description

METHOD, APPARATUS, SYSTEM AND COMPUTER PROGRAM FOR IMAGE RETRIEVAL BASED ON LOCAL COLOR DIFFERENCE BACKGROUND OF THE INVENTION Field of the Invention [0001]

The present invention relates to a method for retrieving an image.

More particularly, the present invention relates to a method and apparatus for efficiently extracting descriptors for local chrominance of an image.

Recently, high quality video data is generated through various mobile devices, and a lot of multimedia data is shared on the web through a social network service.

In this situation, various services using image search are being provided. In order to retrieve images, feature points are extracted from image data. SIFT and CEDD are conventional techniques for extracting feature points.

Patent Registration No. 10-1511050 (published on April 13, 2015)

An object of the present invention is to provide a method for efficiently extracting local color difference information from an image. It is another object of the present invention to provide a method of distributing and storing an image database in which the color difference information is indexed to a corresponding image.

In a server according to an embodiment of the present invention, a method of processing an image includes: collecting a plurality of images and extracting a feature region of the collected image; Extracting pixels of the feature region by skipping at predetermined intervals; Generating data on a change value of a color of adjacent pixels as a descriptor; Merging the descriptors to generate index values, and indexing the index values.

Further, in a server according to an embodiment of the present invention, a method of processing an image comprises: collecting a plurality of images and extracting a feature region of the collected image; Generating color information of a first pixel among pixels of the characteristic region and color information of a second pixel adjacent to the first pixel; If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And generating a descriptor vector of the image by merging a value for the first category and a value calculated for the second category.

Further, in a server according to an embodiment of the present invention, a method of processing an image comprises the steps of: collecting a plurality of images, color information of the collected images, and time information for uploading the images to the database, Creating a database; Receiving a query image, extracting a feature region of the query image; Extracting one value for the first category if the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the first pixel and the color information of the second pixel have different values for the second category; Wherein the first query is a query image descriptor and the second query query image is a query image query. The method of claim 1, further comprising: Indexing the image; And retrieving an image similar to the query image by comparing an index value of the query image with an index value of the database image.

Further, a server for processing an image according to an embodiment of the present invention includes: a crawler module for collecting a plurality of images; Extracts a plurality of images, extracts a feature region of the collected image, extracts pixels of the feature region by skipping at predetermined intervals, and generates data on the change value of the color of the adjacent pixels as a descriptor Parsing module; And an indexer module for merging the descriptors to generate index values, and indexing the index values into the images.

Further, a server for processing an image according to an embodiment of the present invention includes: a crawler module for collecting a plurality of images; And extracting color information of a first pixel of the feature region and color information of a second pixel adjacent to the first pixel, wherein the color information of the first pixel, the color information of the first pixel, If the color information of two pixels has the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel are extracted for the second category Calculating a change value of a color of the first pixel and the second pixel with respect to the second category and merging a value calculated for the first category with a value calculated for the second category, And a parsing module for generating a descriptor vector of the image.

Further, a server for processing an image according to an embodiment of the present invention includes: a database for collecting a plurality of images, indexing color information of the collected images and time information for uploading the images to the database, as index values; A communication unit for receiving a query image; And extracting a feature region of the query image, extracting pixels of the feature region by skipping at predetermined intervals, calculating data of a change value of a color of neighboring extracted pixels as a descriptor of the query image, A vector for a descriptor and time information for receiving the query image are indexed into the query image and an image similar to the query image is searched by comparing an index value of the query image with an index value of the database image And a control unit.

Further, a server for processing an image according to an embodiment of the present invention includes: a database for collecting a plurality of images, indexing color information of the collected images and time information for uploading the images to the database, as index values; A communication unit for receiving a query image; And if the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, And if the color information of the first pixel and the color information of the second pixel have different values for the second category, the first pixel and the second pixel Calculating a change value of a color of a pixel and calculating a value for the first category and a value calculated for the second category as a descriptor of the query image, Index information of the query image is compared with the index value of the database image, And a control unit for searching for an image similar to the image.

Further, in a server according to an embodiment of the present invention, a computer program recorded on a computer-readable storage medium for processing an image includes: a function of collecting a plurality of images and extracting a feature region of the collected image; A function of extracting the pixels of the characteristic region by skipping at preset intervals; A function of generating data on a change value of a color of adjacent pixels extracted by the descriptor; And merging the descriptors to generate an index value and to index the image.

Further, in a server according to an embodiment of the present invention, a computer program recorded on a computer-readable storage medium for processing an image includes: a function of collecting a plurality of images and extracting a feature region of the collected image; A function of generating color information of a first pixel and pixels of a second pixel adjacent to the first pixel in the characteristic region; If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And a function of generating a descriptor vector of the image by merging a value for the first category and a value calculated for the second category.

Further, in a server according to an embodiment of the present invention, in a server, a computer program recorded on a computer-readable storage medium for processing an image collects a plurality of images, and stores color information of the collected images and the image in a database A function of creating a database indexing the uploading time information on the image as an index value; Receiving a query image, extracting a feature region of the query image; Extracting pixels of the feature region by skipping at predetermined intervals, calculating data of a change value of a color of the adjacent pixels extracted as a descriptor of the query image, and receiving a vector for the descriptor and the query image Indexing the query image using time information as an index value; And retrieving an image similar to the query image by comparing the index value of the query image with the index value of the database image.

Further, in a server according to an embodiment of the present invention, in a server, a computer recorded on a computer-readable storage medium for processing an image may be configured to collect a plurality of images, to upload color information of the collected images, The ability to create a database indexing time information into the image as an index value; Receiving a query image, extracting a feature region of the query image; Extracting one value for the first category if the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the first pixel and the color information of the second pixel have different values for the second category; Wherein the first query is a query image descriptor and the second query query image is a query image query. The method of claim 1, further comprising: The ability to index images; And retrieving an image similar to the query image by comparing the index value of the query image with the index value of the database image.

According to the present invention, local color difference descriptors of an image can be efficiently extracted, and processing of a high-quality image can be quickly performed.

Further, according to the present invention, it is possible to construct an image database by adding time information to image information, and in particular, it is possible to shorten the time required for a photo search.

Further, according to the present invention, it is possible to efficiently and accurately search for the same similarity between the image displayed on the client and the image stored in the database of the server. Therefore, you do not need to insert the tag for search into the image in advance, and you can automatically match similar images without reference to separate text.

1 is a flowchart for explaining an image database building method according to an embodiment of the present invention;
2 is a flowchart for explaining an image search method according to an embodiment of the present invention.
3 is a diagram for explaining a method of extracting local color difference descriptors according to an embodiment of the present invention
4 is a flowchart for explaining an example of a similar image search method according to another embodiment of the present invention
5 is a block diagram for illustrating a configuration of a client for transmitting a query image and receiving a similar image according to an embodiment of the present invention
6 is a block diagram illustrating a configuration of a server operating as an image search engine according to an embodiment of the present invention;

It is to be understood that the present invention is not limited to the description of the embodiments described below, and that various modifications may be made without departing from the technical scope of the present invention. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

In the drawings, the same components are denoted by the same reference numerals. And in the accompanying drawings, some of the elements may be exaggerated, omitted or schematically illustrated. It is intended to clearly illustrate the gist of the present invention by omitting unnecessary explanations not related to the gist of the present invention. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Image retrieval can be utilized in various ways. For example, product information search.

For example, a user viewing a blog using a desktop may be curious about the name or price of a baby article displayed on a web page. In this case, according to the embodiment of the present invention, the user can capture the object in the device, query the captured image, access the image search server with the image similar to the baby goods contained in the query and the goods information linked with the similar image have.

To this end, an extension program for generating a query may be installed in a web browsing tool in the form of a plug-in, an ActiveX, a tool bar, or the like, or may be embedded in a specific application in the form of an SDK in a mobile device or may operate as an independent application.

The service server can retrieve an image similar to the image object included in the query in the goods database and provide the search result to the user device. When a user clicks a link linked to a search result, a web page of an online retailer of the baby goods can be displayed as a new tab of the web browser.

Another example is when a user clicks on the icon associated with the search result, the name of the infant's article; Photographs or drawings such as a perspective view, a front view, a back view, a side view, or a plan view; Age of use; How to use; Manufacturer information; Purchase Review; Open market price or lowest price information; Or information about other recommended products that can be recommended due to the interest of the child care products, may be displayed in a pop-up format. Furthermore, online sales pages of KAMSEC products can be displayed.

There are two frameworks for image retrieval that can be used in this way: one is text based image retrieval and the other is content based image retrieval.

In text-based image retrieval, all images in the database are labeled with keywords, and image retrieval can be performed on a label basis. However, there is a problem that the subject of the human being is intervened in the labeling process and the mathematical result can not be derived. On the other hand, content-based image retrieval performs retrieval using visual similarity between images.

It is an object of the present invention to provide a method for processing a visual feature of an image for image retrieval. More specifically, according to the present invention, pixels of a feature region of an arbitrary image can be skipped at predetermined intervals, and data on a change value of a color of adjacent pixels extracted can be generated as a local descriptor.

According to another embodiment of the present invention, if the color information of the first pixel and the color information of the second pixel adjacent to the first pixel in the pixels of the feature region of an arbitrary image have the same value for any category, And if data having different values are to be generated for the category, data for calculating the change values of the colors of the first pixel and the second pixel may be generated as a local descriptor.

An image retrieval method according to an embodiment of the present invention will be described in more detail below with reference to the accompanying drawings.

1 is a flowchart for explaining an image search method according to an embodiment of the present invention.

In step 110, the image search service server may collect images to build an image database.

More specifically, the service server includes a crawler, a parser, and an indexer to collect web documents on-line and access various types of text information linked to images and images included in the web documents.

For example, a crawler might collect a list of Web addresses of interest, track Web sites, and track links to deliver data related to this "G" to the service server, And extracts the image included in the page and the text information linked with the image, and the indexer can index the corresponding position and meaning.

Further, the service server can process the collected images. This is for labeling the object based on the image and further providing the search result based on the similarity of the image, without relying on other text information.

To do this, the service server extracts features of the collected images and indexes feature information of the images for efficiency in searching.

More specifically, the service server can detect the feature region of the images (Interest Point Detection). The feature region refers to a main region for extracting a descriptor, that is, a feature descriptor, of an image for judging whether or not there is a similarity among images. (Step 120)

These characteristic areas are defined by the blobs that are distinguished from the contour lines and corners of the image by the corners of the corners and the surrounding areas, the areas which are unchanged or crooked according to the deformation of the image, And may be targeted to a patch (piece) of an image or an entire image.

According to the embodiment of the present invention, the service server can extract, as the feature region, a region that is invariable to distortion such as a change in viewpoint or environment in the image.

Furthermore, the service server can extract the descriptor from the feature region (Descriptor Extraction). (Step 130) The feature descriptor is a vector value representing the characteristics of the image.

The feature descriptor can be calculated using the position of the feature region for the image, or the brightness, color, sharpness, gradient, scale, or pattern information of the feature region. For example, the feature descriptor may be calculated by converting the brightness value, the brightness value, or the distribution value of the feature region into a vector.

On the other hand, the feature descriptor for the image may include a global descriptor, a frequency descriptor, a binary descriptor, or a neural network descriptor as well as a local descriptor based on the feature region as described above. . ≪ / RTI >

More specifically, the feature descriptor may include a global descriptor (e.g., a global descriptor) for converting the brightness, color, sharpness, gradient, scale, pattern information, etc. of each of the regions divided by the entire image or an arbitrary reference, Global descriptor).

For example, the feature descriptor may include a frequency descriptor for converting the number of times the specific descriptors classified into the image are included in the image, the number of inclusions of the global feature such as the previously defined color table, etc. into a vector value, A binary descriptor which extracts whether the size of each element value constituting the descriptor is larger or smaller than a specific value and transforms it into an integer type and uses it in a layer of a neural network Or a neural network descriptor for extracting image information used for classification.

In particular, the service server according to the embodiment of the present invention can extract the color information of the feature region by a descriptor. More specifically, the service server can generate a descriptor using the color difference value between pixels constituting the feature region.

The descriptor may be generated by expressing a color difference, a chrominance difference, or a color difference between neighboring pixels among the pixels constituting the characteristic region of the image as a vector value. In this case, when the color difference is the same, when the adjacent pixel is larger than the reference pixel, and when the adjacent pixel is smaller than the reference pixel, the color difference may be expressed by three kinds of values.

However, according to the above-described method, the number of pixels constituting the feature region of the image multiplied by 3 is required. However, there is a problem that the resources consumed here are too large to be suitable for high-quality image processing.

According to the embodiment of the present invention, a chrominance descriptor of an image can be generated for pixels extracted while skipping at predetermined intervals, without generating for all the pixels of the feature region. Since the color information of adjacent pixels is often overlapped in the high-quality data, even if the pixels are extracted at an arbitrary interval and only the color difference information of the extracted pixels is expressed by the descriptor, the efficiency of the search is not lowered.

In this case, the service server extracts the pixels of the feature region by skipping at predetermined intervals, and generates data on the change values of the colors of the adjacent pixels as a descriptor.

According to another embodiment of the present invention, a chrominance descriptor of an image may be generated by extracting only one value when the color information of an arbitrary pixel is the same as the color information of an adjacent pixel, and calculating a color difference value when the color information is different. According to this method, chrominance information can be represented by the descriptor together with the color information of all the pixels, so that there is an effect that more precise searching is possible.

Furthermore, the service server can integrate the extracted local descriptors and express them as a feature information vector, and index the corresponding images as an index value. (Step 140)

At this time, the service server according to the embodiment of the present invention may generate the index value by combining the feature information vector and the time information of the image. The time information of the image may illustrate the time the image was created, the time the image was received at the service server, the time the image was uploaded to the database, and so on.

According to this method, the index value includes not only the local information of the image but also the time information of the image, so that it is possible to provide a search service considering time variables.

Meanwhile, the service server according to the embodiment of the present invention can construct an image database in a distributed environment. At this time, the service server can apply the hash function to divide the database and store the divided database on the designated machine.

At this time, the service server according to the embodiment of the present invention may generate the hash function by combining the feature information vector and the time information of the image. The time information of the image may illustrate the time the image was created, the time the image was received at the service server, the time the image was uploaded to the database, and so on.

FIG. 2 is a flowchart illustrating a method of retrieving an image similar to a query image in a service server according to an embodiment of the present invention.

In step 210, the service server may build an image database. An image database building method according to an embodiment of the present invention is the same as the one described in the description of FIG.

In step 220, the service server may receive a query from the client to query the image.

To do this, the client can run a program to generate queries and receive results. The program may be an extension program installed in a mobile terminal or a web browsing tool of a desktop in the form of a plug-in, an active X, or the like, or may be an application installed in a mobile terminal.

The program can be downloaded and installed from a service server according to an embodiment of the present invention, and further downloaded from a service server through a website server such as an online application market.

When the service server receives the query, it can identify the image included in the query and pre-process the image for later retrieval. Unlike subsequent image processing, the pre-processing of the query image can change the scale of the query image itself for search, resize, or change the resolution according to a predetermined format.

Further, in step 230, the service server may extract the color information from the query image as a descriptor.

More specifically, the service server can first detect the feature region of the query image. The feature region refers to a main region for extracting a descriptor, that is, a feature descriptor, of an image for judging whether or not there is a similarity between images.

These characteristic areas are defined by the blobs that are distinguished from the contour lines and corners of the image by the corners of the corners and the surrounding areas, the areas which are unchanged or crooked according to the deformation of the image, And may be targeted to a patch (piece) of an image or an entire image.

According to the embodiment of the present invention, the service server can extract, as the feature region, a region that is invariable to distortion such as a change in viewpoint or environment in the image.

Further, the service server can extract the feature descriptor from the feature region. The feature descriptor is a vector value representing the characteristics of the image.

The feature descriptor can be calculated using the position of the feature region for the image, or the brightness, color, sharpness, gradient, scale, or pattern information of the feature region. For example, the feature descriptor may be calculated by converting the brightness value, the brightness value, or the distribution value of the feature region into a vector.

On the other hand, the feature descriptor for the image may include a global descriptor, a frequency descriptor, a binary descriptor, or a neural network descriptor as well as a local descriptor based on the feature region as described above. . ≪ / RTI >

More specifically, the feature descriptor may include a global descriptor (e.g., a global descriptor) for converting the brightness, color, sharpness, gradient, scale, pattern information, etc. of each of the regions divided by the entire image or an arbitrary reference, Global descriptor).

For example, the feature descriptor may include a frequency descriptor for converting the number of times the specific descriptors classified into the image are included in the image, the number of inclusions of the global feature such as the previously defined color table, etc. into a vector value, A binary descriptor which extracts whether the size of each element value constituting the descriptor is larger or smaller than a specific value and transforms it into an integer type and uses it in a layer of a neural network Or a neural network descriptor for extracting image information used for classification.

In particular, the service server according to the embodiment of the present invention can extract the color information of the feature region from the query image as a descriptor. More specifically, the service server can generate a descriptor using the color difference value between pixels constituting the feature region.

The descriptor may be generated by expressing a color difference, a chrominance difference, or a color difference between neighboring pixels among the pixels constituting the characteristic region of the image as a vector value. In this case, when the color difference is the same, when the adjacent pixel is larger than the reference pixel, and when the adjacent pixel is smaller than the reference pixel, the color difference may be expressed by three kinds of values.

However, according to the above-described method, the number of pixels constituting the feature region of the image multiplied by 3 is required. However, there is a problem that the resources consumed here are too large to be suitable for high-quality image processing.

According to the embodiment of the present invention, a chrominance descriptor of an image can be generated for pixels extracted while skipping at predetermined intervals, without generating for all the pixels of the feature region. Since the color information of adjacent pixels is often overlapped in the high-quality data, even if the pixels are extracted at an arbitrary interval and only the color difference information of the extracted pixels is expressed by the descriptor, the efficiency of the search is not lowered.

In this case, the service server extracts the pixels of the feature region by skipping at predetermined intervals, and generates data on the change values of the colors of the adjacent pixels as a descriptor.

According to another embodiment of the present invention, a chrominance descriptor of an image may be generated by extracting only one value when the color information of an arbitrary pixel is the same as the color information of an adjacent pixel, and calculating a color difference value when the color information is different. According to this method, chrominance information can be represented by the descriptor together with the color information of all the pixels, so that there is an effect that more precise searching is possible.

Furthermore, the service server can integrate the extracted local descriptors into a feature information vector and index the query image with the index value. (Step 240)

At this time, the service server according to the embodiment of the present invention may combine the feature information vector with the time information of the query image to generate an index value for the query image. The time information of the image may illustrate the time at which the query image was generated, the time at which the query image was received at the service server, and so on.

According to this method, since the index value includes not only the local information of the query image but also the time information of the image, a search service considering the time variable can be provided.

In step 250, the service server can calculate the similarity between the query image and the database image using the index value of the query image and the index value of the database image.

For example, the service server determines whether the distance between the feature descriptor vector value of the database image and the feature descriptor vector value of the query image is within a preset threshold value, and if the distance is within the threshold value, It can be judged to be similar.

Further, the service server can re-verify the search result. The re-verification determines the degree of matching between the query image and the database image by using the attribute of the feature information such as color-difference information, scale, rotation information and distribution information of the feature descriptor for the image, and when the degree of matching exceeds a predetermined threshold value It can proceed in a form that passes re-verification.

3 is a diagram for explaining a local color difference descriptor extracting method according to an embodiment of the present invention.

When color information of an image is extracted by a descriptor, a descriptor can be generated by using a color difference value between pixels constituting the feature region. The descriptor may be generated by expressing a color difference, a chrominance difference, or a color difference between neighboring pixels among the pixels constituting the characteristic region of the image as a vector value. In this case, when the color difference is the same, when the adjacent pixel is larger than the reference pixel, and when the adjacent pixel is smaller than the reference pixel, the color difference may be expressed by three kinds of values.

However, according to the above method, the number of pixels constituting the feature region of the image multiplied by 3 is required. FIG. 3A exemplifies this, and in this case, a large amount of resources are required to represent the descriptor, which is not suitable for high-quality image processing.

According to the embodiment of the present invention, the chrominance descriptor of the image can be generated by extracting only one value when the color information of an arbitrary pixel is the same as the color information of an adjacent pixel, and calculating the color difference value when the color information is different. FIG. 3B illustrates this. According to this method, chrominance information together with color information of all pixels can be expressed by a descriptor. Therefore, more precise searching is possible, and redundant data is eliminated, which is efficient.

4 is a flowchart illustrating an example of an image retrieval method according to an embodiment of the present invention.

The contents of step 410, step 420, step 455, and step 460 of FIG. 4 are substantially the same as those of FIG. 1 or FIG. 2 described above.

In step 425, the client can proceed to distinguish objects included in the displayed image. For example, a client can distinguish objects such as a background, a person, a coffee can, and a lotion using the outline of the displayed image.

In step 430, the client may receive a user selection input for the separated image object. At this time, the client can set the object such as the background or the person to be selected by the user according to the purpose of the implementation, and to select only the object for the merchandise available for sale.

In step 440, the client may perform a process of generating feature information on the object image selected by the user as a query image. More specifically, the client can detect the feature region of the image object selected by the user and extract the feature descriptor. The concrete contents are the same as those described in the description of FIG. 1 or FIG. On the other hand, the number of characteristic regions and the number of characteristic descriptors in the query image can be set differently for each client considering the performance of the process.

In step 445, the client can transmit the feature information of the generated query image to the server.

According to the embodiment of FIG. 4, not only the image displayed on the client, but only the feature information of the object image selected by the user can be moved through the network. Therefore, the amount of data moved to the network can be greatly reduced, and the time required for uploading and downloading is reduced, thereby improving the speed of the entire system.

On the other hand, in step 465, the service server can perform the process of converting the received query image feature information according to the same criterion as that of classifying the feature information of the database image.

The contents of steps 490 and 440 are substantially the same as those of FIG. 1 or FIG. At step 450, the client can display the search results.

5 is a block diagram illustrating a configuration of a client according to an embodiment of the present invention. Although a mobile terminal such as a smart phone, a tablet, and a PDA is described in this specification, the present invention is not limited thereto, and a client according to an embodiment of the present invention may be a personal computer or a notebook computer.

5, the client 500 according to the embodiment of the present invention includes a communication unit 510, a touch screen unit 530, an input unit 520, a storage unit 540, and a control unit 560 And may further include an audio processing unit and a video processing unit, not shown in FIG.

The communication unit 510 performs a function of transmitting and receiving corresponding data for wired / wireless communication of the client 500. The communication unit 510 receives data through a wired / wireless channel and outputs the data to the control unit 550, And transmits the data through the channel.

The touch screen unit 530 includes a touch panel 533 and a display unit 536. In particular, the touch panel 533 according to the embodiment of the present invention may receive a user's selection input of a product object among the images displayed on the display unit 536, and the display unit 536 may display the product information .

The input unit 520 receives a user's input for controlling the client 500, generates an input signal, and transmits the input signal to the controller 550. In the preferred embodiment of the present invention, the client 500 is fully operable only with the touch screen unit 530. [ In this case, the touch panel 533 can perform the function of the input unit 520.

The storage unit 540 stores programs and data necessary for the operation of the client 500, and can be divided into a program area and a data area. The data area of the storage unit 540 according to an embodiment of the present invention may store feature information of a query image or a query image. Furthermore, the program area of the storage unit 540 according to the embodiment of the present invention may store an application for causing the client to execute the process of displaying the product information.

The application includes a function of transmitting the image displayed on the display unit 536 to the server as a query image, a function of receiving product information identical or similar to the query image from the server, and a function of displaying the product information Can be performed.

The control unit 550 controls the overall operation of each component of the client. In particular, the control unit 550 of the client according to the exemplary embodiment of the present invention transmits the image displayed on the display unit 536 to the server as a query image, and transmits a search result, which is the same as or similar to the query image, 510 and controls the display unit 536 to display the search result.

6 is a block diagram illustrating a configuration of a server 600 according to an embodiment of the present invention. 6, the server 600 according to the embodiment of the present invention may include a communication unit 610, a storage unit 630, and a control unit 620. The server 600 may include an input unit And may further include a display unit.

6, the server 600 is illustrated as including a communication unit 610, a storage unit 630, and a control unit 620, but each block may exist physically and separately. For example, the storage unit 630 physically exists in a database (not shown) and may be connected to the controller 620 of the server 600 through the communication unit 710.

The communication unit 610 transmits and receives data for wired / wireless communication of the server 600. The communication unit 610 receives data through a wired / wireless channel and outputs the data to the control unit 620. The control unit 620 transmits the data output from the control unit 620 to a non- Through the network. In particular, the communication unit 610 according to the embodiment of the present invention can perform a function of receiving a query image from a client.

The storage unit 630 stores programs and data necessary for the operation of the server 600, and can be divided into a program area and a data area. In the data area of the storage unit 630 according to an embodiment of the present invention, the storage unit 630 may store an image database, text information associated with the image, and query image feature information.

Further, the program area of the storage unit 640 according to the embodiment of the present invention may store a computer program that causes the server to execute a process of providing an image search result to a client.

The computer program includes a function of generating a database for a plurality of images and extracting feature information of an image included in the database, a function of receiving a displayed image from a client as a query image from the client, A function of searching for feature information of the database image to determine a matching image matching feature information of the query image, and a function of providing the client with information about the matching image.

The control unit 620 controls the overall operation of each component of the server 700. In particular, the control unit 620 according to the embodiment of the present invention generates a database for a plurality of images, extracts feature information of the images included in the database, searches the feature information of the database image, And may control the communication unit 610 to provide information on the search result to the client

The embodiments of the present invention disclosed in the present specification and drawings are merely illustrative of specific embodiments of the present invention and are not intended to limit the scope of the present invention in order to facilitate understanding of the present invention. It is to be understood by those skilled in the art that other modifications based on the technical idea of the present invention are possible in addition to the embodiments disclosed herein.

500: Client
510: communication unit of the client
520: Input of the client
530: Client's touch screen
540:
550: Control section of the client
600: Server
610: Communication section of the server
620:
630: Storage section of the server

Claims (20)

A method for processing an image in a server,
Collecting a plurality of images and extracting feature regions of the collected images;
Extracting pixels of the feature region by skipping at predetermined intervals;
Generating color information of the extracted first pixels and second pixels extracted by the interval from the first pixels;
If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And
Merging a value for the first category and a value for the second category to generate a descriptor vector of the image; And
Generating index values by combining the descriptor vectors of the images with time information of the images generated, and indexing the index values to the image. 2. The method of claim 1, wherein after generating the descriptor vector,
Dividing the database into hash values by using time information for uploading the images to the database and storing location information of the images, and distributing and storing the divided databases. A method for processing an image in a server,
Collecting a plurality of images and extracting feature regions of the collected images;
Generating color information of a first pixel among pixels of the characteristic region and color information of a second pixel adjacent to the first pixel;
If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And
Merging a value for the first category and a value for the second category to generate a descriptor vector of the image. ≪ Desc / Clms Page number 22 > 4. The method of claim 3, wherein after generating the descriptor vector,
Indexing the descriptor vector into the image as an index value; And
Dividing the database into hash values by using time information for uploading the images to the database and storing location information of the images, and distributing and storing the divided databases. 5. The method of claim 4,
Receiving a query image, generating a descriptor for a color change value of pixels constituting the feature region of the query image;
Generating a feature information vector by combining the descriptors, generating an index value by combining the feature information vector with the time information of the query image, calculating an index value of the query image and an index value of the image included in the database Calculating a distance of the object;
Determining that the image included in the database and the query image are similar if the distance is less than or equal to a preset threshold value. A method for processing an image in a server,
Generating a database in which a plurality of images are collected and color information of the collected images and time information for uploading the images to the database are indexed to the images by index values;
Receiving a query image, extracting a feature region of the query image;
Extracting pixels of the feature region by skipping at predetermined intervals, calculating data of a change value of a color of the adjacent pixels extracted as a descriptor of the query image, and receiving a vector for the descriptor and the query image Indexing the query image using time information as an index value; And
And comparing the index value of the query image with an index value of an image included in the database to search for an image similar to the query image. 7. The method of claim 6, further comprising: after the step of creating the database,
Dividing the database into hash values of the time information and the storage location information of the image, and distributing and storing the divided databases. 8. The method of claim 7,
Calculating a distance between an index value of the query image and an index value of an image included in the database; And
Determining that the image included in the database and the query image are similar if the distance is less than or equal to a preset threshold value. A method for processing an image in a server,
Generating a database in which a plurality of images are collected and color information of the collected images and time information for uploading the images to the database are indexed to the images by index values;
Receiving a query image, extracting a feature region of the query image;
Extracting one value for the first category if the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the first pixel and the color information of the second pixel have different values for the second category;
Wherein the first query is a query image descriptor and the second query query image is a query image query. The method of claim 1, further comprising: Indexing the image; And
And comparing the index value of the query image with an index value of an image included in the database to search for an image similar to the query image. 10. The method of claim 9, wherein after the step of creating the database,
Dividing the database into hash values of the time information and storage location information of the image, and distributing and storing the divided databases. 11. The method of claim 10,
Calculating a distance between an index value of the query image and an index value of an image included in the database; And
Determining that the image included in the database and the query image are similar if the distance is less than or equal to a preset threshold value. A server for processing an image,
A crawler module for collecting a plurality of images;
Extracts a plurality of images, extracts a feature region of the collected image, extracts pixels of the feature region by skipping at predetermined intervals, and generates data on the change value of the color of the adjacent pixels as a descriptor Parsing module; And
An indexer module for merging the descriptors to generate index values, and indexing the images;
The parsing module comprises:
Wherein the color information of the first pixel and the color information of the second pixel are generated in a first pixel and a second pixel extracted in the first pixel, Wherein if the color information of the first pixel and the color information of the second pixel have different values for the second category, For each category, a change value of a color of the first pixel and a second pixel, merging a value for the first category and a value calculated for the second category to generate a descriptor vector of the image,
Wherein the indexer module generates the index value by combining the descriptor time vector of the image with the descriptor vector of the image.
delete A server for processing an image,
A crawler module for collecting a plurality of images; And
The color information of the first pixel and the color information of the second pixel adjacent to the first pixel are generated based on the color information of the first pixel and the color information of the second pixel, If the color information of the pixel has the same value for the first category, extracts one value for the first category, and if the color information of the first pixel and the color information of the second pixel are different for the second category Calculating a change value of a color of the first pixel and the second pixel for the second category and merging a value for the first category and a value calculated for the second category, And a parsing module for generating a descriptor vector for the descriptor. A server for processing an image,
A database for collecting a plurality of images, indexing color information of the collected images and time information for uploading the images to the database as index values;
A communication unit for receiving a query image; And
Extracting a feature region of the query image, extracting pixels of the feature region by skipping at predetermined intervals, calculating data of a change value of a color of adjacent pixels extracted as a descriptor of the query image, And an index value of the query image is compared with an index value of the image included in the database to calculate an image similar to the query image, And a controller for searching the server. A server for processing an image,
A database for collecting a plurality of images, indexing color information of the collected images and time information for uploading the images to the database as index values;
A communication unit for receiving a query image; And
If the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, And if the color information of the first pixel and the color information of the second pixel have different values for the second category, the first pixel and the second pixel Calculating a change value of the color of the query image, and calculating a value for the first category and a value calculated for the second category as a descriptor of the query image, Time information as an index value, indexes the query image, and compares the index value of the query image with the index value of the image included in the database, The server comprising a control unit that searches for images similar to the query image. A computer program recorded on a computer-readable storage medium for processing an image in a server,
Collecting a plurality of images and extracting feature regions of the collected images;
A function of extracting the pixels of the characteristic region by skipping at preset intervals;
A function of generating color information of a first pixel extracted and a second pixel extracted by the interval from the first pixel;
If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And
A function of generating a descriptor vector of the image by merging a value for the first category and a value calculated for the second category; generating an index value by combining the time information of the image with the descriptor vector of the image; And performs a function of indexing the image. A computer program recorded on a computer-readable storage medium for processing an image in a server,
Collecting a plurality of images and extracting feature regions of the collected images;
A function of generating color information of a first pixel and pixels of a second pixel adjacent to the first pixel in the characteristic region;
If the color information of the first pixel and the color information of the second pixel have the same value for the first category, one value is extracted for the first category, and the color information of the first pixel and the color information of the second pixel Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the pixel has a different value for the second category; And
And a function of generating a descriptor vector of the image by merging a value for the first category and a value calculated for the second category. A computer program recorded on a computer-readable storage medium for processing an image, in a server, in a server,
A function of collecting a plurality of images, generating a database in which color information of the collected images and time information of uploading the images to the database are indexed into the images by index values;
Receiving a query image, extracting a feature region of the query image;
Extracting pixels of the feature region by skipping at predetermined intervals, calculating data of a change value of a color of the adjacent pixels extracted as a descriptor of the query image, and receiving a vector for the descriptor and the query image Indexing the query image using time information as an index value; And
And a function of searching an image similar to the query image by comparing an index value of the query image with an index value of an image included in the database. A computer program recorded on a computer-readable storage medium for processing an image, in a server, in a server,
A function of collecting a plurality of images, generating a database in which color information of the collected images and time information of uploading the images to the database are indexed into the images by index values;
Receiving a query image, extracting a feature region of the query image;
Extracting one value for the first category if the color information of the first pixel of the feature region and the color information of the second pixel adjacent to the first pixel have the same value for the first category, Calculating a change value of the color of the first pixel and the second pixel with respect to the second category if the color information of the first pixel and the color information of the second pixel have different values for the second category;
Wherein the first query is a query image descriptor and the second query query image is a query image query. The method of claim 1, further comprising: The ability to index images; And
And a function of searching an image similar to the query image by comparing an index value of the query image with an index value of an image included in the database.

Images