JP3010450B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing system for generating an image having a corresponding color from a given sentence.

[0002]

2. Description of the Related Art As a conventional apparatus of this kind, there is one disclosed in, for example, "Knowledge Engineering and Artificial Intelligence", 56-59 pp. 65-72. FIG. 2 shows a configuration diagram of this conventional image retrieval system. In FIG. 2, reference numeral 21 denotes inference means for inferring an input sentence, 22 denotes skeleton description means for describing an image, and 23 denotes inference means 21. Image interpretation knowledge means 24 for giving knowledge, 24 is an image file for storing a target image, and 25 is an image description means for storing the contents of the image file 24.

[0005] The conventional image retrieval system configured as described above operates as follows. Various images are recorded in the image file means 24 in advance. Then, when the user wants to obtain an image corresponding to, for example, an image including a person, an image of a house with a square window, or an image of a sunny day, as an image search request, When "people" is input from a keyboard or the like, an image file with the heading "people" is searched. When "house", "window", and "square" are input, the tree structure described in the image description means 25 is traced, and an image file corresponding to "house with a square window" is searched.

[0004]

However, in the above configuration, an image cannot be generated for an arbitrary given natural language, and a target image is always stored in an image file in advance. There is a need. as a result,
It is necessary to make the storage capacity of the image file 24 enormous.

[0005] In view of the problems of the conventional image retrieval system, the present invention provides a method of inputting a character string in a natural language.
It is an object of the present invention to provide an image processing device that freely generates an image corresponding to the contents.

[0006]

In order to achieve the above object, the present invention provides a morphological analysis means for analyzing a morpheme of an input character string, a morphological dictionary for storing morphemes to be subjected to the analysis, At least a noun about a morpheme
And a syntactic analysis means for analyzing adjective information and its nouns
An image dictionary that outputs the corresponding image and its analyzed form
Parameter corresponding to the color of the adjective based on the color information
Output the color table and the contents of the color table
Multiplying the output of the image dictionary by a parameter
Multiplication means for converting the corresponding image to the corresponding color
An image processing apparatus comprising:

Further, according to the present invention, the input character string is used for speech recognition.
The image processing apparatus described above, wherein the image processing apparatus is inputted from a means.
It is.

[0008]

According to the present invention, a morphological analysis unit analyzes a morpheme of an input character string using a morphological dictionary storing morphemes, and a syntactic analysis unit analyzes at least noun and adjective information on the morpheme. Then, the image dictionary outputs an image corresponding to the noun, the color table outputs a parameter corresponding to the corresponding color based on the color information among the analyzed adjectives, and the multiplication means outputs the content of the color table. Therefore, the image is converted into the corresponding color by multiplying the output of the image dictionary by a parameter.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an image processing apparatus according to an embodiment of the present invention. In FIG. 1, a morphological analysis unit 2 is a unit for performing a morphological analysis on an input sentence 1. The morphological dictionary 3 is a dictionary for storing morphemes to be subjected to the morphological analysis, and the morphological analysis means 2 performs morphological analysis by matching the morphemes stored in the morphological dictionary 3 to determine the contents of the text. . At this time, the morphological dictionary 3 also stores grammatical information such as whether the corresponding morpheme is a noun, a verb, an adjective, an adverb, etc. In particular, the morpheme indicates a color with respect to the adjective. If it is, a flag indicating that it is color information is stored. Parsing means 4
Is determined as a noun as a result of the morphological analysis by the morphological analysis means 2, referring to the noun table 7 to detect the address corresponding to the noun, and the image dictionary 8 reads the corresponding image corresponding to the noun from the corresponding address. Is output. Further, as a result of the morphological analysis, when it is determined that the adjective is an adjective representing a color, the color table 5 is referred to and a coefficient relating to the color is extracted. The coefficient multiplier 6 multiplies each of the red, green, and blue primary color outputs of the image data detected from the image dictionary 8 by the extracted count of the color table 5, colors the corresponding color, and colors the corresponding image. Is colored in the color required by the input sentence and output to the output signal line 9.

The operation of the embodiment of the image processing apparatus configured as described above will be described below in accordance with the actual processing flow.

As an example, suppose that the following sentence is input and a request to create an image corresponding to the following sentence is made.

"The red apple and the blue tangerine are on the desk." At this time, the morphological dictionary 3 stores the independent words and the auxiliary words extracted from the Japanese dictionary. Independent words such as "red""i""apple""to""blue"
Includes the attached words "te", "ni", "no", "wo", "ga",
The input sentences are compared in order from the independent word, and the matched one is regarded as the analysis result. When a plurality of words are matched, the longest independent word is used as the matching result.

Next, matching is performed on the portion where the independent word is omitted by using the attached word to determine the attached word. Next, a check is made as to whether the attached word attribute and the independent word can be combined with respect to the attached word attribute. As a result, if the pair of the independent word and the auxiliary word that can be connected can be determined, the parsing is terminated.
Perform the same operation for the candidate adjunct.

As a result, "red""i""apple""to"
"Blue", "I", "Tangerine", etc. are output.

The parsing means 4 extracts "apple", "tangerine" and "desk" determined as nouns and "red" and "blue" determined as color information.

In the noun table 7, the correspondence between each noun and the address of the image dictionary 8 is stored in advance, and images corresponding to "apple" and "tangerine" are extracted from the image dictionary 8. If the information determined to be color is “red” from the color table 5, a parameter of (red, blue, green) = (1, 0, 0)
If it is "blue", the parameter of (red, blue, green) = (0, 1, 0) is extracted from the color table 5 according to the mixing ratio of another color if it is another color.
The multiplying unit 6 outputs the images (red,
(Blue, green). By outputting the sum as an image output to the output signal line 9, an image of "red apple" and "blue tangerine" can be generated. When no color is specified as in the case of “desk”, it is natural to output an image without color conversion by setting (red, blue, green) = (1, 1, 1).

As described above, according to this embodiment, by providing the morphological analysis means 2, the noun table 7, the color table 5, and the multiplier 6, an image corresponding to an input natural language can be obtained. Can be generated.

If a well-known voice recognition device is provided in the input portion, a sentence expressed in voice can be converted into an image.

In this embodiment, the components can be processed programmatically using a computer.

[0021]

As described above, according to the present invention, a morphological analysis means for analyzing a morpheme of an input character string, a morphological dictionary for storing morphemes to be analyzed, and an image corresponding to the morpheme are stored. Since the image corresponding to the morpheme analyzed by the morphological analysis means is extracted from the image dictionary, an image can be generated from an arbitrary sentence, and the practical effect is large.

Further, by using a color table, colors can be easily added in a natural language relating to colors.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a block diagram illustrating a conventional image search system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input character string 2 Morphological analysis means 3 Morphological dictionary 4 Syntax analysis means 5 Color table 6 Multiplier 7 Noun table 8 Image dictionary

──────────────────────────────────────────────────続 き Continuation of the front page (56) References IEICE Technical Report HC90-12 “Basic system for media conversion communication integrating voice and image language”, The Institute of Electronics, Information and Communication Engineers, July 1990 10th (58) Fields investigated (Int.Cl. ⁷ , DB name) G06T 11/00-11/80 G06F 17/30

Claims (2)

(57) [Claims] 1. A morphological analyzer for analyzing a morpheme of an input character string, a morphological dictionary for storing morphemes to be analyzed, a syntactic analyzer for analyzing at least noun and adjective information on the morpheme, An image dictionary that outputs an image corresponding to the noun, a color table that outputs a parameter corresponding to the color based on the color information among the analyzed adjectives, and an image dictionary that outputs the image dictionary according to the contents of the color table. An image processing apparatus comprising: multiplying means for converting an output image into a corresponding color by multiplying an output by a parameter. 2. The image processing apparatus according to claim 1, wherein said input character string is input from a voice recognition unit.