JPH0830643A - Image retrieval device - Google Patents

Abstract

PURPOSE:To easily judge the necessity/unnecessity of an image and to quickly and surely execute image retrieval by a method wherein to each divided image data is added a sub-keyword that makes the contents of the divided image data recognizable, and the divided image data and information correspondingly to the sub-keyword are stored. CONSTITUTION:Image managing information including an image information name, a keyword, divided image data, and a sub-keyword corresponding to the divided image data is alreadyly stored in a storage means 1, and when retrieving information, e.g. a keyword, is inputted from a retrieving condition input means 2, image managing information corresponding to the keyword is read out by an image information retrieving means 3. When retrieving information, e.g. an image information name and a sub-keyword, is inputted from the means 2, only a divided image of a part corresponding to the sub-keyword out of the image of the inputted image information name is read out from the means 1 by a partial image extracting means 5 and can be displayed on a retrieved result displaying means 4. Thereby only the divided image data of a part to be observed can be observed and the retrieving time of image data can be shortened.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image retrieval device,
In particular, the present invention relates to an image search device that searches for image information using a keyword.

[0002]

2. Description of the Related Art A conventional image retrieval apparatus extracts, for example, data of one original image, image data of a specific portion, data indicating the position and size of the specific portion with respect to the whole, and the specific portion from the original image. It is divided into image data of the remaining images that have been removed, and these data are stored in an image storage device in an individually accessible state, and only the image of the specific portion is displayed during image search, and the necessity of the image is eliminated. With respect to an image of a specific portion that the operator judges to be necessary, after moving the image to a correct position with respect to the entire image, the remaining image data is read from the image storage device so that the entire image is displayed. The technique is disclosed in Japanese Patent Laid-Open No. 4-280372 (hereinafter, referred to as first prior art).

Further, when storing one image data, it is divided into a plurality of rectangular areas, each rectangular area is given a rectangular area data size, a vertical and horizontal size, and an address, and the image data is displayed. By storing the image data corresponding to the display order of, by displaying only a part of the stored image data,
Japanese Patent Laid-Open No. 4-209071 (hereinafter, referred to as a second prior art) is a technique for shortening the display time of the image data and enabling a high-speed search when the plurality of image data are roughly viewed.
Is disclosed in.

[0004]

However, in the first prior art, when the image of the specific portion is registered, the subjectivity of the operator is included. Therefore, when another operator performs image retrieval, Due to the difference in the operator's subjectivity, it may not be possible to determine whether the original image is necessary or not even by looking at the display of the image of the specific portion, and there is a problem that the search takes time. Further, there is a problem that the operator cannot know what kind of image is drawn unless the image of the specific portion is actually displayed.

In the second prior art, the display order of the images in the respective rectangular areas is predetermined when the image data is stored, and therefore the display order cannot be changed during the image search. Therefore, for example, if there are four rectangular areas, even if it is desired to display from the image of the third rectangular area,
There is also a problem that it takes time to search because the images must be displayed in order from the second image. Further, since it is not known what image of which rectangular area is stored, there is a problem that the image of the first rectangular area must be displayed. Further, when an operator different from the operator who registered the rectangular area searches for an image as in the case of the first prior art, the feature of the displayed image is different due to the difference in subjectivity of each operator. Therefore, there is a problem that the necessity of the image cannot be sufficiently determined.

An object of the present invention is to eliminate the above-mentioned problems of the prior art, divide one image information into arbitrary sizes,
An object of the present invention is to provide an image search device that searches for image information using a keyword that can be recognized to some extent without displaying the divided image information and that can display the divided image information in any order.

[0007]

In order to achieve the above object, the invention of claim 1 stores image management information including an image information name, a keyword, divided image data and a sub-keyword corresponding to the divided image data. Storage means, search information input means for inputting search information such as image information name, keyword and subkeyword, and image information for searching image management information corresponding to the search information input from the search information input means Search means, search result display means for displaying the predetermined image management information extracted from the storage means by the image information search means, and partial image data corresponding to the search information input from the search information input means. It is characterized in that it is provided with a partial image extracting means for extracting from the storage means and outputting to the search result display means.

The invention of claim 2 is characterized in that an arbitrary sub-keyword can be input as search information from the search information input means.

[0009]

According to the first aspect of the invention, the image management information including the image information name, the keyword, the divided image data and the sub-keyword corresponding to the divided image data is stored in the storage means, and the search information input means is used. When search information, for example, a keyword is input, the image information search means reads the image management information corresponding to the keyword from the storage means. Then, when the search information, for example, the image information name and the sub-keyword is input to the search information input means,
Only the divided image of the portion corresponding to the sub-keyword of the image of the input image information name can be read from the storage means by the partial image extraction means and displayed on the search result display means. Therefore, only the divided image data of the desired portion can be viewed, and the search time for the image data can be shortened.

According to the second aspect of the present invention, since any sub-keyword can be input as the search information from the search information input means, it is possible to display the divided image data of a desired portion in any order. Become.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the image retrieval apparatus of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a functional block diagram showing a schematic configuration of an embodiment of the present invention. In the figure, 1 is image information, a name of the image information, a keyword,
And storage means 2 for storing image management information such as sub-keywords, and 2 is search information input means for inputting search information such as the image information name, keywords and sub-keywords, such as a keyboard and a mouse. .
Reference numeral 3 denotes an image information retrieval means for retrieving image information corresponding to a keyword inputted from the retrieval information input means 2, and 4 denotes a keyword inputted to the image information retrieval means 3 extracted from the storage means 1. The search result display means 5 for displaying the image information, the image information name, and the sub-keyword retrieves the partial image information corresponding to the image information name and the sub-keyword from the storage means 1 and outputs it to the search result display means 4. It is a partial image extracting means.

FIG. 2 is a block diagram showing a hardware configuration for registering and retrieving image information. In the figure,
Reference numeral 11 is a central processing unit (hereinafter, CPU) that controls the system of the image retrieval apparatus, 12 is a memory in which a control program for operating the CPU 11 is stored, 13 is a scanner for reading an original image, and 14 is, for example, from the CPU 11. It is a scanner control device for controlling the scanner 13 by an output signal. The CPU 11 includes the functions of the image information searching means 3 and the partial image extracting means 5. 15
Is a magnetic disk device for storing information, 16 is a magnetic disk control device for controlling the magnetic disk device 15, and 17 is a VRAM for expanding a display image as image data. Reference numeral 18 is a mouse and 19 is a keyboard, which correspond to the search information input means 2 in FIG. Reference numeral 20 is a display device for displaying image data and character data such as commands, which corresponds to the search result display means 4, 21 is a display control device for controlling the display device 20, and 22 is a large amount of image management information. An optical disk device 23 for storing is an optical disk control device for controlling the optical disk device 22. Reference numeral 24 is a bus for electrically connecting the above-described components, such as an address bus, a data bus, and a control bus.

FIG. 3 is an explanatory diagram showing an example of a rectangular area obtained by dividing one original image into an arbitrary size and division number.
FIG. 4 is a conceptual diagram of a specific example of the image management information. In the figure, 31 is an image data management unit which stores the image information name of the original image of FIG. 3, vertical and horizontal sizes of the original image data, a color lookup table, and the like, and 32 is added to the original image information. A keyword storage area for storing the generated keywords, 33 is a division number storage area for storing the number of divisions when the image information of one original image is divided into several rectangular areas, and 34 to 39 are each divided rectangles. Area information, for example, pointers of divided image data ~, starting point coordinates (x, y), size (h, w), and sub-keywords "sea" to "tree" added to the divided image data ~ are rectangles. Data storage area stored by area, 40 to 4
Reference numeral 5 denotes a divided image data storage unit for storing divided image data 1 to 3 for each rectangular area.

Next, the operation of the image information registration processing according to the embodiment of the present invention will be described with reference to FIG. FIG. 5 is a flowchart illustrating the operation of the image information registration process.
In step S1, the original image is read by the scanner 13 and input as image information into the image search device. In step S2, the image information name for the image information is input by the keyboard 19, and in step S3, the keyword is input.

In step S4, image division information (for example, pointer, starting point coordinates (x = 0, y = 0) and size of divided image data for dividing the image information into arbitrary sizes by the mouse 18 or the keyboard 19 is used. (H = 3
0, w = 50)) is input, and in step S5, the sub-keyword "sea" for the divided image data is input.

In step S6, it is determined whether or not the division of the image information is completed. When the desired division work is completed and the sub-keywords are added to all the divided image data items ~, the process proceeds to step S7. In step S7, the image information name input in steps S1 to S5 is the image data management unit 31. And the keyword is the keyword storage area 3
2, divided image data ~ origin coordinates (x, y), size (h, w), and sub-keywords "sea" to "tree"
In the data storage areas 34 to 39 together with the pointer
A numerical value (= 6) corresponding to the number of divisions of the divided image data is stored in the number-of-divisions storage area 33, and divided image data to are stored in the management table of FIG. 4 in correspondence with the divided image data storage units 40 to 45, respectively. It The management table is created in the optical disk device 22.

As described above, the image data of the original image read by the scanner 3 is divided into rectangular divided image data having an arbitrary size and an arbitrary number of divisions, and the sub-keyword corresponding to each divided image data. It can be stored together with the optical disk device 22.

Next, the operation of the image information search processing according to the embodiment of the present invention will be described with reference to the drawings. FIG. 6 is a flowchart illustrating the operation of the image information search process.
In step S11, the keyword of the desired image information is input by the search information input means 2 and step S1
In 2, the image information to which the same keyword is added based on the keyword is searched by the image information searching means 3, and in step S13, the image information name of the image information based on the keyword and the sub-keywords "sea" to "". wood"
Is displayed on the search result display means 4. In step S14, the image information name of the desired image information, the sub-keyword,
For example, “river” is input from the search information input means 2. Then, the partial image extraction means 5 accesses the storage means 1, and in step S15, the divided image data corresponding to the input image information name and the sub-keyword "kawa" is displayed on the search result display means 4.

In step S16, it is determined whether or not the other divided image data, ... Are displayed. If the divided image data displayed in step S15 is the desired divided image data, the process proceeds to step S17. , If another divided image data, ... Is desired to be displayed, the operation from step S14 is repeated. In step S17, it is determined whether or not the operation of the image information search process is completed,
When the desired image information is obtained, the process ends. If the image information is not the desired image information or if another image information is desired to be retrieved for other reasons, the operation from step S11 is repeated.

As described above, by arbitrarily selecting the sub-keyword displayed on the display 20, the divided image data corresponding to the selected sub-keyword can be displayed on the display 20.

In this embodiment, as shown in FIG.
Although the case where the image management information of one original image is collectively stored in a continuous storage area has been described, each data of the image management information may be stored in separate storage areas.

[0022]

As is apparent from the above description, according to the invention of claim 1, the image data of the stored image is divided into a plurality of rectangular divided image data having an arbitrary size, and the divided image data is divided. A sub-keyword that allows the content to be recognized is added to each data, and the divided image data and information are stored so as to correspond to the sub-keyword. The divided image data of the image corresponding to the sub-keyword can be displayed only by reading the information and subsequently inputting the sub-keyword in the image information.

Even if the operators who perform the image information registration process and the image information search process are different from each other, the contents of the image to be searched for can be recognized from the sub-keywords, so that it is easy to properly judge whether the image is necessary or not. Therefore, it becomes possible to perform a quick and accurate image search.

According to the second aspect of the present invention, the divided images can be arbitrarily displayed regardless of the order of the image information registration processing.

[Brief description of drawings]

FIG. 1 is a block diagram showing functions of an embodiment of the present invention.

FIG. 2 is a block diagram showing a hardware configuration of an embodiment of the present invention.

FIG. 3 is an explanatory diagram showing an example of original image information divided into arbitrary sizes and division numbers.

FIG. 4 is a conceptual diagram of a specific example of image management information.

FIG. 5 is a flowchart illustrating an operation of image information registration processing.

FIG. 6 is a flowchart illustrating an operation of image information search processing.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Storage means, 2 ... Search information input means, 3 ... Image information search means, 4 ... Search result display means, 5 ... Partial image extraction means, 11 ... Central processing unit, 12 ... Memory, 13 ... Scanner, 14 ... Scanner Control device, 15 ... Magnetic disk device, 16 ... Magnetic disk control device, 17 ... VRAM, 1
8 ... Mouse, 19 ... Keyboard, 20 ... Display device, 21 ... Display control device, 22 ... Optical disc device, 23 ... Optical disc control device, 24 ... Bus, 31 ... Image information name, 32 ... Keyword storage area, 33 ... Division number Storage area, 34-39 ... Data storage area, 40-45 ... Divided image data storage section.

Claims (2)

[Claims] 1. An image search device for inputting image search information and displaying a search result, wherein image management information including an image information name, a keyword, divided image data and a sub-keyword corresponding to the divided image data. A storage means for storing the search information, a search information input means for inputting search information such as an image information name, a keyword and a sub-keyword, and image search information corresponding to the search information input from the search information input means. Image information search means, search result display means for displaying the predetermined image management information extracted from the storage means by the image information search means, and a partial image corresponding to the search information input from the search information input means Partial image extracting means for extracting data from the storage means and outputting the data to the search result display means. Image search apparatus. 2. The image search apparatus according to claim 1, wherein any sub-keyword can be input as search information from the search information input means.