JP2002304628A - System and method for information retrieval and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a client server system for information retrieval to retrieve an image with a less load on a server and good retrieval efficiency. SOLUTION: A client side calculates the feature quantity of an inputted image (S2), calculates the image similarity according to the calculated feature quantity of the image and feature quantities of a plurality of stored images, and determines a retrieval object image according to the calculated image similarity (S3 to S5) to request details of the retrieval object image from a server (S6). The server side once receives the request sent from the client (S11), it reads image data corresponding to the request and sends them to the client (S12).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a client-server type information retrieval system for retrieving a desired image from an image database storing a plurality of images, a method thereof, and a storage medium.

[0002]

2. Description of the Related Art Various retrieval methods have been proposed for retrieving a desired image from a database storing a large number of image data. These include, for example, a method of associating non-image information such as a keyword and a photographing date and time with an image and performing a search based on the image. There are two main methods.

In the latter case, an image is presented,
There is a method of searching for an image stored in a database using the feature amount of the image as a search key, and this is called a similar image search. This method has an advantage that a user who does not have special knowledge about image processing can be provided with an intuitive and easy-to-understand search interface.

[0004]

Since it is not efficient for a plurality of users to separately hold a huge image database, various client / server type image databases for centrally managing image data have been proposed and put into practical use. However, when performing the above-described similar image search, the load on the server that manages the database is large. Therefore, there is a problem that the cost of the server increases when trying to realize a server that satisfies the specifications. Was.

[0005] Further, when an image of a search result is transmitted from a server to a client via a network, the network bandwidth is hindered and it takes time to display an image on the client side, which hinders efficient image search. Had become.

The present invention has been made in view of the above-mentioned conventional example, and stores image feature amounts required for searching for similar images on the client side, searches for similar images on the client, and searches for desired images found by the search. An information retrieval system and method capable of requesting an image from a client to a server and transmitting the image data from the server to the client to reduce the load on the server and perform an image search with high search efficiency. The purpose is to provide.

[0007]

To achieve the above object, an information retrieval system according to the present invention has the following arrangement. That is, a client-server type information retrieval system that retrieves a desired image from an image database in which a plurality of images are stored, wherein the client stores a feature amount of each of the plurality of images; An image input unit for inputting an image, a feature amount calculation unit for calculating a feature amount of the image input by the image input unit, the feature amount obtained by the feature amount calculation unit, and the first storage unit. A similarity calculating unit that calculates an image similarity based on the stored image feature amount; an image determining unit that determines a search target image based on the image similarity calculated by the similarity calculating unit; A first transmitting unit for requesting the server for details of the search target image determined by the image determining unit, and a first transmitting unit for receiving image data transmitted from the server in response to the request A receiving unit, wherein the server stores at least image data of the plurality of images, and a second storing unit for receiving the request transmitted by the first transmitting unit of the client. It has a receiving unit and a second transmitting unit for reading out image data corresponding to the request from the second storing unit and transmitting the image data to the client.

[0008] In order to achieve the above object, the information search method of the present invention comprises the following steps. That is, an information retrieval method in a client-server type information retrieval system for retrieving a desired image from an image database storing a plurality of images, wherein the client calculates a feature amount of the input image, and Image similarity is calculated based on the calculated image feature amount and the stored feature amounts of each of the plurality of images, a search target image is determined based on the calculated image similarity degree, and the search is performed. The server requests the details of the target image from the server, and upon receiving the request transmitted from the client, the server reads out the image data corresponding to the request from the storage unit that stores the image data of at least a plurality of images. And transmits the data to the client.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing a schematic configuration of a client-server type information processing system according to an embodiment of the present invention.

In FIG. 1, reference numerals 1a to 1d denote client computers, which have the same configuration, and for the sake of explanation, only the configuration of the client computer 1a is shown in detail here. Reference numeral 11 denotes a server computer which is connected to these client computers 1a to 1d via a network 12, and can communicate with each other. The configuration of the server computer 11 is the same as that of the client computer. FIG. 1 shows a case where the number of client computers is four, but an arbitrary number of client computers can be generally connected. Also, the number of server computers 11 is not limited to one, and an arbitrary number of server computers can be connected. For example, there may be a configuration in which a plurality of server computers manage a database having the same contents for different reasons, such as using different server computers for each database or being geographically separated.

Since the configuration of the computer network 12 is known, its description is omitted.
Is a LAN (local area network), a so-called intranet, WAN (wide area network)
Then, the present embodiment can be realized on the so-called Internet.

Next, the configuration of the client computer (same for the server) will be described.

In FIG. 1, reference numeral 101 denotes a CPU which controls the entire client computer. 1
A keyboard 02 is operated by the user together with the mouse 102a, and is used to input various data and instructions. A display unit 103 includes a CRT, a liquid crystal, and the like. A ROM 104 and a RAM 105 constitute a storage unit of the computer, and store programs executed by the client computer and various data used by the computer. Reference numeral 106 denotes a hard disk, and 107 denotes a floppy (registered trademark) disk, which constitutes an external storage device used for a file system of the client computer. 108 is a printer. A network interface 109 is connected to the network 12 via the network interface.

FIG. 2 is a diagram showing an example of an operation screen displayed on the display unit 103 of the client computer when executing the search processing.

Reference numeral 21 denotes a user drawing area, in which an arbitrary image input according to a user's instruction is displayed. 22, a mouse cursor; 23, a color designation scroll bar; 24, a clear button; and 25, a search execution button.

In 26a to 26h, icons of the images hit by the search are displayed. 27 is a next candidate display button, and 28 is a processing end button.

The user can write an illustration resembling an image to be searched in the user drawing area 21 by using the drawing tool realized by software stored in the RAM 105. The outline of the operation by software when drawing such an illustration is as follows.

Reference numeral 23 denotes a scroll bar for designating a color of an image drawn by the mouse cursor 22. R, G, and B color components can be designated in order from the top. That is, when the bar is moved to the left, the density of the color component is reduced, and when the bar is moved to the right, the density is specified to be higher. When the clear button 24 is pressed, the image drawn in the drawing area 21 is erased, and the whole is painted white. The user can draw a free image on the drawing area 21 by moving the mouse cursor 22 using the pointing device 102a.

After the desired image is drawn in the drawing area 21 and the search execution button 25 is designated, an image search process based on the image drawn in the drawing area 21 is executed.

FIG. 3 is a flowchart showing the general flow of image retrieval processing in the client-server type system according to this embodiment. In the figure, server processes S11 and S12 are server computers 11
The client processes S1 to S10 are executed by any of the client computers 1a to 1d.

First, the client process will be described.

In step S1, the user draws an illustration on the display unit 103 in the drawing area 21 so as to resemble a desired image. Next, the process proceeds to step S2, and the feature amount of the image drawn in the drawing area 21 is calculated. This may be instructed by the search start button 25 as described later, or may be performed at predetermined time intervals. Next, the process proceeds to step S3, and a similar image is searched based on the obtained feature amount. Next, the process proceeds to step S4, and the similar image obtained by the search is displayed on the display unit 103. Then, the process proceeds to step S5, and the operator selects a desired image from the icons displayed in the list. Next, the process proceeds to step S6, where the ID of the image is transferred to the server computer 11 via the network 12.

Next, the process proceeds to step S7, waits for the image data sent from the server 11, and when the image data is received, the process proceeds to step S8. In step S8, the display unit 10
3 displays a detailed image. Next, the process proceeds to step S9, in which it is determined whether or not the selection and display work is to be ended. If the selection and display work is to be ended, the process proceeds to step S10, and it is determined whether to end the search work. If the search is to be continued, the process returns to step S1.

Next, the processing in the server computer 11 will be described.

First, in step S11, it is determined whether or not an image request from a client computer has been received. If not, the process returns to step S11. When the image ID for requesting the image is received, step S12 is performed.
Then, the requested image data is transmitted to the client computer 11 that has requested the image, and the process returns to step S11.

With the above configuration, a similar image search system in a client / server format is configured.

Hereinafter, each of the above steps will be described in detail. (1) “Description of Step S1” In step S1, an illustration is drawn in the drawing area 21 shown in FIG. 2 by a user operation, and the process proceeds to the next step S2 at an appropriate timing. This processing will be described with reference to the flowchart shown in FIG. (X0, y0) is a variable that stores the position indicated by the mouse cursor 22 immediately before,
(X1, y1) is a variable in which the current position of the mouse cursor 22 is stored.

First, in step S21, it is determined whether or not the mouse 102a has been moved by the user. If it has not been moved, the process returns to step S21. That is, mouse 1
02a forms a loop for monitoring whether the mouse cursor 22 has been moved. When the mouse 102a is moved, the process proceeds to step S22, and it is determined whether or not the button of the mouse 102a is pressed. Position (x
(1, y1) is substituted for (x0, y0), and the process returns to step S21. Thus, only the movement of the mouse cursor 22 is performed without performing drawing.

On the other hand, if the mouse button is depressed in step S22, that is, if the mouse button is being dragged, step S2
Proceed to 3. Here, the position (x0, y0) of the immediately preceding mouse cursor 22 and the current mouse cursor position (x1, y0)
During 1), a straight line is drawn in the color designated by the color designation scroll bar 23 (Line (x0, y0)-(x1, y1)). Next, at step S24, the current mouse pointer position (x1,
y1) is substituted for (x0, y0).

Thus, the processing in step S1 ends.

According to such a process, from the user's point of view, every time a little drawing is added to the illustration in the drawing area 21, it seems that the search process based on the illustration is automatically executed. Note that the search process may be executed when the search button 25 is instructed after the drawing of the illustration in the drawing area 21 is completed, depending on the user's preference or the like. (2) “Explanation of Step S2 (Calculation of Image Feature Amount)” In step S2, the image feature amount of the illustration drawn in the drawing area 21 is calculated.

As shown in FIG. 5, the size of the drawing area 21 is W pixels in the horizontal direction and H pixels in the vertical direction. This is divided into three parts in the horizontal direction and two parts in the vertical direction, that is, a total of six parts, and the area (0, 0) and the area (1,
0),... Area (2, 1). R of each of these areas,
The average value of the G and B values is calculated, and the image feature amount of the drawn illustration is obtained based on a total of 18 (= 6 × 3) numerical values.

FIG. 6 is a flowchart showing a process for obtaining such image feature values. With reference to FIG. 6, the flow of the image feature amount calculation process according to the present embodiment will be described.

First, in step S31, a variable K is initialized to "0". Next, in step S32, the variable j is set to “0”,
In step S33, the variable i is initialized to “0”. Here, the variable K is for indicating the RGB average value (array d) in each area of FIG.
Are variables for indicating each area in FIG.

Then, the process proceeds to a step S34, wherein the average value of the R component of the area (i, j) is substituted for the K-th element d (K) of the array d. Also, the average value of the G component is given by d (K + 1),
Then, the average value of the B component is substituted for d (K + 2). The method of calculating the average value of these R, G, B components is shown in FIG.
It will be described later with reference to the flowchart of FIG.

Then, the process proceeds to a step S35, wherein the variable K is increased by "3". In step S36, the variable i is increased by “1”, and the attention area is shifted to the area on the right in the drawing area 21. Step S37
To determine whether the value of the variable i is greater than "2".
If it is larger than "2", the process proceeds to step S38 assuming that the processing of the upper one-line region has been completed, but otherwise, the process returns to step S34 to execute the above-described process.

In step S37, when i> 2 and the area of the first row ends, the process proceeds to step S38, where
The variable j is increased by “1”. Then, the process proceeds to a step S39, wherein the variable j is compared with “1”.
The processing is completed on the assumption that the processing for the area on the line is completed, but if not, the flow returns to step S33 to execute the above processing.

Thus, when the processing shown in the flowchart of FIG. 6 is completed, the arrays d (0) to 18
The image feature amount of the illustration image drawn in the drawing area 31 is stored in d (17).

In this case, the image is divided into six rectangular areas having the same area in order to calculate the characteristic amount. However, the division is not limited to the rectangular area, and may be an area having a more complicated shape. The number of divided areas may be increased. When the number of regions is increased, the number of elements of the image feature amount is not 18 but, of course, increases or decreases accordingly.

FIG. 7 shows R, G, B for each area shown in FIG.
It is a flowchart explaining the calculation process of the average value of a component.

The image data is represented by R (X, Y), G (X,
Y) and B (X, Y). However, 0 ≦ X <W, 0 ≦ Y <H, and the upper left corner of the image is the starting point (0, 0).

In the following processing, X0 ≦ X <X1, Y0 ≦
The average density of the partial area where Y <Y1 is calculated, and variables DR, D
It is assumed that the average densities of R, G, and B are stored in G and DB, respectively, and return.

An area corresponding to each area (i, j) is represented by an upper left corner (X0, Y0) and a lower right corner (X1, Y1), and each is defined as follows.

X0 = W × i / 3, X1 = W × (i + 1) / 3 Y0 = H × j / 2, Y1 = H × (j + 1) / 2 Therefore, the constants X0, X1, Y0, and Y1 are calculated as described above. After the initialization, the processing shown in the flowchart of FIG. 7 is executed.

First, in step S41, variables DR, DG,
DB and C are all initialized to “0”. Next, step S
42, the variable Y is initialized to "Y0", and the
At 43, the variable X is initialized with "X0". Next, the process proceeds to step S44, where R (X, Y) is added to the variable DR. Similarly, the variable DG has G (X, Y) and the variable DB has B (X, Y).
Y) is added. Also, the counter C for counting the number of pixels is incremented by +1.
I do.

Next, in step S45, the variable X is incremented by "1". Next, in step S46, the variable X is compared with "X1". Since the process has been completed, the process proceeds to step S47,
If not, the process returns to step S44, and the process for the next pixel is performed.

Thus, in step S46, one of the areas
When the process for the pixels in the row is completed, the process proceeds to step S47, in which the variable Y is increased by "1". In step S48, the variable Y is compared with "Y1". Then, the process proceeds to step S49; otherwise, the process returns to step S43 to execute the above-described processing.

In step S49, variables DR, DG, D
B is divided by the number of pixels C in the area. Each of the variables DR, DG, and DB thus obtained is an average agricultural degree of RGB obtained by dividing the sum of the pixel densities in the area by the number of pixels.

In this way, the average RGB density of each area shown in FIG. 5 is obtained. (3) “Explanation of Step S3 (Similar Image Search)” In step S3 of FIG. 3, similar image search is performed based on the feature amount of the image.

It is assumed that N pieces of image data are stored in the hard disk 106, and the feature amount of each image is calculated and stored in advance by the same method as described above. Image data is well-known JPEG, FlashP
ix or a standard file format such as RDBMS (relational database
Management system) in a unique format.

The image feature amount is 2 × N × 18.
Dimensional array D (n, i) (where 0 ≦ n <N, 0 ≦ i <1
8).

At this time, the inter-image distance S (n) between the presentation image and the n-th image is defined by the following equation.

S (n) = Σ (D (n, i) −d (i)) ² Here, it is determined that the smaller the inter-image distance S (n), the higher the similarity of the images.

First, the inter-image distance S (n) (0 ≦ n <N) between all the N images and the presentation image is calculated. M (0 <
A similar image is searched for by selecting M <N).

FIG. 8 is a flowchart showing a process for obtaining the inter-image distance S (n).

First, in step S51, a variable n is initialized to "0". Then, the process proceeds to a step S52, wherein the variables i, S
(N) is initialized to "0". And step S53
Where, the two-dimensional array D (n, i) and the RGB concentration d (i)
Is added to S (n). Next, step S5
Then, the process proceeds to step S55, where the variable i is increased by one, and
Then, the variable i is compared with "18", and if they are equal, it is determined that one row of the two-dimensional array has been completed, and the process returns to step S56 otherwise. In step S56, the variable n is increased by "1", Te, and the process proceeds to step S57, where the variables n and N are compared.
Otherwise, the process returns to step S52.

By executing the above processing, the inter-image distance between the presentation image and all the stored images is stored in the two-dimensional array S (n).

Next, the process of selecting M images in ascending order of the image distance S (n) and storing the image numbers in the array T () will be described with reference to FIG.

FIG. 9 is a flowchart showing a process for extracting M items from the image-to-image distance S (n) in ascending order of distance. At the start of this process, the variable N and the M to be extracted are equal to each other. It shall be given in advance.

First, in step S61, a variable j is initialized to "0", and in step S62, a variable i is initialized to "0". Next, in step S63, the variable min is set to “0”.
Then, the variable L is initialized with a sufficiently large value ().

Next, the process proceeds to step S64, where S (i) and L
And if S (i) <L (this condition is always satisfied at first), the process proceeds to step S65; otherwise, the process proceeds to step S66. In step S65, the value of “i” is substituted for the variable min, and the value of “S (i)” is substituted for the variable L. Then, in a step S66, the variable i is increased by "1". Then, the process proceeds to step S67, where the variable i is compared with the value of “N”. If the values are equal, the process proceeds to step S68; otherwise, the process returns to step S64. Step S
At 68, the value of "min" is substituted into the array T (j).
Then, the process proceeds to step S69, in which a sufficiently large value (∞) is substituted for S (min). Then, in a step S70, the variable j is increased by "1". Then, in a step S71, the variable j is compared with the value of “M”. If the values are equal, the process is completed, and if not, the process proceeds to a step S62.

When the processing of FIG. 9 is executed, the array T (j)
In (0 ≦ j <M), the image numbers are stored in the order of the degree of similarity to the presentation image (from the smallest image distance S (n)). (4) "Description of Step S4 (Similar Image List Display)" The processing will be described with reference to the operation screen of FIG.

In the areas 26a to 26h in FIG. 2, the similar images searched by the above processing are displayed in a reduced size (icons or thumbnail images). Here, in the area 26a,
The image corresponding to the image number T (0) having the highest similarity, the area 26b displays the image corresponding to the image number T (1) having the next highest similarity,... Displays the image with the lowest similarity. The displayed icon or thumbnail image data is stored in the hard disk 106 of the client computer in advance, and is decoded and displayed.

Then, in FIG. 2, a "Next" button 2
7, the next candidate, ie, the next highest similarity T
Icons of the images corresponding to (8) to T (15) are reduced and displayed in the areas 26a to 26h. By repeating this, up to the last similar image number T (M-1) can be displayed.

Further, each of the reduced images in the areas 26a to 26h is "selected" by using the pointing device 102a.
Be prepared to do it. (5) “Explanation of Step S5 (Image Selection)” In step S5, the operator selects an icon corresponding to a desired image from the plurality of icons displayed in the areas 26a to 26h using the pointing device 102a. I do. (6) “Explanation of Step S6 (Detailed Image Request)” In Step S6, communication is performed between the client computer and the server computer 11, and Step S6 is performed.
The image ID selected in step 5 is transmitted to the server computer 11. (7) “Explanation of Step S7” In Step S7, it is monitored whether image data has been transmitted from the server computer 11. Upon receiving the image data from the server computer 11, step S8
Proceed to. (8) “Explanation of Step S8” In Step S8, the image data selected in Step S5 is displayed in detail in Step S7.

FIG. 11 shows a display unit 103 for displaying details of a selected image on a client computer.
FIG. 3 is a diagram showing an example of a screen configuration.

Reference numeral 111 denotes an image display area in which a selected image is displayed. By operating the reduce button 113 and the enlarge button 114, a part of the image can be enlarged and displayed,
The image may be confirmed by, for example, reducing the display. A character display area 112 is an image display area 111.
Displays the metadata associated with the image displayed in. The metadata includes, for example, shooting date and time and copyright information. This metadata is stored in step S7.
It may be embedded in the image data received from the server computer 11, may be embedded in the icon image data stored in the hard disk 106 of the client computer, or may be embedded in the hard disk 10 in association with the image information of the client computer.
6 may be stored in another area.

Here, when the close button 115 is instructed, the detailed display ends, and the flow advances to step S9. In step S9, the operator selects another image to be displayed, if any, and repeats the operation of displaying the detailed image. (9) “Explanation of Step S9 (Selection / Display)” In Step S9, it is determined whether or not to end the selection / display processing. If the processing is to be continued, the process returns to Step S5, and another image can be selected. . (10) "Explanation of Step S10 (End of Search Work)" In Step S10, the processing completion button 28 (Exi
It is determined whether or not t) has been instructed, and if so, the process is completed. If not instructed, the process returns to step S1, and the operator can continue drawing the illustration. (11) “Explanation of Step S11 (Detailed Image Request Reception)” Step S11 is executed by the server computer 11, and here, it is monitored whether an image request is sent from a client computer. Upon receiving the image ID requested from the client computer, the process proceeds to step S12. (12) “Explanation of Step S12 (Image Data Transmission)” In step S12, communication is performed between the client computer and the server computer 11, and the image data corresponding to the received image ID is transmitted to the client computer that has requested it. Send. And step S11
And waits for the next image request sent from the client computer.

FIG. 2 shows a case where the number of client processes is one. However, since each of a plurality of client computers executes a client process,
In step S11, image requests from all client computers are monitored, and accordingly, in step S12
It is necessary to transmit the requested image.

Further, a plurality of server processes may be executed in parallel on the server computer 11 to simultaneously respond to requests from a plurality of clients.

[Other Embodiments] When an image content vendor such as a stock photo vendor is a server and a customer is a client, it is necessary to charge for each content. In such a case, online image sales can be realized by exchanging authentication information and billing information in step S6 in FIG.

At this time, the image feature amount and the icon data group required on the client side are stored in a CD-ROM or the like by the image content distributor and delivered to the customer in advance (before the search is actually performed). . The CD-ROM of the icon data may be sold to the customer for a fee or may be provided free of charge. Further, it may be distributed to an unspecified number of potential customers by attaching it to a magazine or the like.

Conventional stock photo vendors sometimes distribute not only printed catalogs but also CD-ROMs as "samples" of their own image contents.
The image search system according to the present embodiment can be realized by naturally extending such a custom.

At the same time, by embedding a unique ID for each CD-ROM medium, it can be used as a key for specifying a user. In the conventional charging system on a network, there is a security problem because the user's credit card number needs to be sent on the network. In this embodiment, however, the icon data CD-ROM is simply used for access. Need only be inserted into the CD-ROM drive of the client computer.

If the CD-ROM media or the like cannot be sent to the client, the image feature amount and the icon data group may be transmitted from the server computer 11 to each client computer via the network before performing the image search. However, it is not always necessary to use transmission means other than the network.

Further, the transmission of the detailed image data from the server computer 11 to the client computer is performed simply by transmitting the data, but is not limited to this. For example, a known IIP (Int
(Ethernet Imaging Protocol), an enlarged display of an image can be performed more efficiently (with less network load). For this purpose, it is only necessary to replace the processing in step S8 in FIG. 2 with the processing in the IIP client and the processing in step S12 with the processing in the IIP server.

Even if the present invention is applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), a device including one device (for example, a copying machine, a facsimile machine, etc.) ) May be applied.

Further, an object of the present invention is to supply a storage medium (or a recording medium) in which a program code of software for realizing the functions of the above-described embodiments is recorded to a system or an apparatus, and a computer (a computer) of the system or the apparatus. Alternatively, the present invention is also achieved when a CPU or an MPU reads and executes a program code stored in a storage medium. In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention. When the computer executes the readout program codes, not only the functions of the above-described embodiments are realized, but also an operating system (OS) running on the computer based on the instructions of the program codes. This also includes a case where some or all of the actual processing is performed and the functions of the above-described embodiments are realized by the processing.

Further, after the program code read from the storage medium is written in a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the program code is read based on the instruction of the program code. , The CPU provided in the function expansion card or the function expansion unit performs part or all of the actual processing,
The case where the function of the above-described embodiment is realized by the processing is also included.

As described above, according to the present embodiment, it is possible to provide a similar image search system with low cost and high search efficiency on a client-server type image database.

[0082]

As described above, according to the present invention, the image feature amount required for searching for a similar image is stored in the client side, the client searches for the similar image, and the desired image found by the search is retrieved. By requesting an image from the client to the server and transmitting the image data from the server to the client, there is an effect that the load on the server can be reduced and an image search with high search efficiency can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a client-server type information processing system according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration example of a display screen in a client computer.

FIG. 3 is a flowchart illustrating a flow of interactive image presentation in the client-server type information processing system according to the embodiment of the present invention.

FIG. 4 is a flowchart illustrating a drawing process (S1) using a mouse in a client computer.

FIG. 5 is a diagram illustrating an example of image division for extracting a feature amount of an image.

FIG. 6 is a flowchart illustrating a flow of a feature amount extraction process (S2) of an image.

FIG. 7 is a flowchart illustrating a process of calculating an average value of R, G, and B components for each region illustrated in FIG. 5;

FIG. 8 is a flowchart showing a process for obtaining an inter-image distance S (n).

FIG. 9 is a flowchart showing a process of extracting M images from the image-to-image distance S (n) in ascending order of distance;

FIG. 10 is a diagram illustrating an example of a screen configuration displaying a detailed image on a client computer.

[Explanation of symbols]

 101 CPU 102 Keyboard 102a Mouse 103 Display 104 ROM 105 RAM 106 Hard Disk 107 Floppy Disk 108 Printer 109 Network Interface (I / F) 11 Server Computer 12 Network 1a to 1d Client Computer

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) G06T 1/00 200 G06T 1/00 200E F term (reference) 5B050 AA08 BA15 CA07 CA08 EA04 EA18 FA02 FA05 FA12 FA13 FA19 GA08 5B075 ND06 NK06 NK31 NK39 PP03 PP13 PQ46 PQ48 PR06 QM05 5L096 AA02 AA06 BA08 CA22 FA32 GA40 HA08 JA03 JA11 KA09

Claims (26)

[Claims] 1. A client-server type information retrieval system for retrieving a desired image from an image database storing a plurality of images, wherein the client stores a feature amount of each of the plurality of images. Means, image input means for inputting an image, feature quantity calculation means for calculating feature quantities of the image input by the image input means, the feature quantity obtained by the feature quantity calculation means, the first A similarity calculating unit that calculates an image similarity based on the feature amount of the image stored in the storage unit; and an image determining unit that determines a search target image based on the image similarity calculated by the similarity calculating unit. First transmission means for requesting the server for details of the search target image determined by the image determination means; and receiving image data transmitted from the server in response to the request. A first receiving unit for storing the image data of at least the plurality of images; and a server for receiving the request transmitted by the first transmitting unit of the client. An information retrieval system, comprising: a second receiving unit for reading the image data corresponding to the request from the second storage unit and transmitting the image data to the client. 2. The information according to claim 1, wherein the client further includes a third storage unit configured to store image data of the icons of the plurality of images in association with a feature amount of each image. Search system. 3. The information retrieval system according to claim 2, wherein said client further comprises an image display means for displaying a candidate image list as an icon based on said image similarity. 4. The information search system according to claim 3, wherein the image determination unit determines an image selected by an operator from the candidate image list as the search target image. 5. The information retrieval system according to claim 1, wherein the server charges the image data when transmitting the image data to the client. 6. The image processing apparatus according to claim 1, wherein the second storage unit stores the image data in a flash memory.
The information retrieval system according to claim 1, wherein the information is stored in an hpix format. 7. The first receiving means and the second transmitting means,
2. The information retrieval system according to claim 1, wherein the information is transmitted and received by IIP (Intemet Imaging Protocol). 8. The server further comprises a conversion unit for converting the format of the image data. When transmitting the image data to the client, the server converts the image data by the conversion unit and transmits the converted image data. 2. The information retrieval system according to claim 1, wherein 9. The information retrieval system according to claim 2, wherein said third storage means is a removable storage medium. 10. The information retrieval system according to claim 9, wherein a fee is charged for distributing the removable storage medium. 11. The removable storage medium includes a CD-ROM.
The information retrieval system according to claim 9, comprising a ROM to a DVD-ROM. 12. The information retrieval system according to claim 9, wherein said removable storage medium includes a memory card. 13. The information retrieval system according to claim 11, wherein the removable storage medium has a copyright protection function for restricting reading and copying of data. 14. The storage medium according to claim 9, wherein the removable storage medium has an ID for specifying an operator of the client, and the server charges the image data based on the ID. An information search system according to any one of the preceding claims. 15. The method according to claim 1, wherein the image data of the icons of the plurality of images and the feature amount of each image data are transmitted from the server to the client in advance and stored in the client before executing a search. Claim 2
Information retrieval system described in. 16. The information retrieval system according to claim 1, wherein said image input means inputs image data by a handwriting operation of an operator. 17. An information retrieval method in a client-server type information retrieval system for retrieving a desired image from an image database storing a plurality of images, wherein a client calculates a feature amount of the input image. Then, an image similarity is calculated based on the calculated image feature and the feature of each of the stored plurality of images, and a search target image is determined based on the calculated image similarity. Requesting the details of the search target image from the server, and upon receiving the request transmitted from the client, the server stores the image data corresponding to the request at least as image data of a plurality of images. An information retrieval method, wherein the information is read from a means and transmitted to the client. 18. The information search method according to claim 17, wherein the client further stores image data of icons of the plurality of images in association with a feature amount of each image. 19. The information search method according to claim 18, wherein the client displays a candidate image list as an icon based on the image similarity. 20. The information search method according to claim 19, wherein the client determines an image selected by an operator from the candidate image list as the search target image. 21. The information search method according to claim 17, wherein the server charges the image data when transmitting the image data to the client. 22. The server according to claim 14, wherein the image data is Flashpix
2. The information is stored in a format.
7. The information search method according to 7. 23. The client and the server may be an IIP
18. The information search method according to claim 17, wherein the information is transmitted and received. 24. The information search method according to claim 17, wherein the server converts the image data and transmits the image data when transmitting the requested image data to the client. 25. The image data of the icons of the plurality of images and a feature amount of each image data are transmitted from the server to the client in advance and stored in the client prior to execution of a search. Claim 1
8. The information search method according to 8. 26. A computer-readable storage medium storing a program for executing the information search method according to claim 17. Description: