JP2007148516A - Image processor, image processing method and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically select an image appropriate to a keyword contained in a sentence input by a user. <P>SOLUTION: This image processor comprises: an image storage part storing images in conformation to keywords showing image contents; an image keyword weight storage part storing weights of keywords; an image keyword extraction part extracting a keyword stored by the image storage part from an input sentence; an image extraction part extracting an image stored by the image storage part; a weight calculation part totaling the weights of keywords stored by the image storage part in conformation to images and stored by the image keyword weight part for each image; and an image selection part selecting images of a predetermined number from an image having a large weight total value, or images larger in total weight value than a predetermined value. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program. In particular, the present invention relates to an image processing apparatus, an image processing method, and an image processing program that extract an image corresponding to a keyword included in an input sentence.

There has been proposed a document creation device for easily selecting an image suitable for a character string of a document (see, for example, Patent Document 1). In the invention described in Japanese Patent Application Laid-Open No. 2004-228561, when creating a sentence in which an image is inserted, first, an attribute of a character string input by a user is determined. Then, a plurality of images that match the character string attribute input by the user are extracted from the images previously associated with the character string attributes. Subsequently, the extracted plurality of images are displayed in a list, and the user selects an image that the user desires to use.
JP 2000-172688 A

  In the invention described in Patent Document 1, since an image is extracted based on the attribute of a character string included in a sentence, a plurality of images corresponding to the sentence and the character string can be extracted. However, when an image is extracted from the character string attribute, if the user does not properly input a character string corresponding to the image that the user is recalling, the image that the user is actually recalling May not be extracted properly.

  Accordingly, an object of the present invention is to provide an image processing apparatus, an image processing method, and an image processing program that can solve the above-described problems. This object is achieved by a combination of features described in the independent claims. The dependent claims define further advantageous specific examples of the present invention.

  In order to solve the above-described problem, in the first embodiment of the present invention, an image processing apparatus stores an image storage unit that stores an image in association with a keyword indicating the content of the image, and an image storage unit. An image keyword weight storage unit that stores the weight of the keyword in association with a keyword that is stored, an image keyword extraction unit that extracts a keyword stored in the image storage unit from newly input text, and an image keyword extraction unit An image extraction unit that extracts an image stored in the image storage unit in association with at least one keyword extracted by the image storage unit, and a keyword stored in the image storage unit in association with the image extracted by the image extraction unit. The keyword weight stored in the image keyword weight storage unit is associated with the same keyword as the keyword extracted by the image keyword extraction unit. A weight calculation unit that calculates a total value for each image extracted by the image extraction unit, and a predetermined number of images or a weight calculation unit that calculates from the image having a large total weight value calculated by the weight calculation unit And an image selection unit that selects an image having a total value of weights larger than a predetermined value from the images extracted by the image extraction unit.

  Moreover, you may further provide the image output part which outputs the image which the image selection part selected with the input sentence. The image storage unit may store images captured by the user in the past. The image processing apparatus further includes a combination weight storage unit that stores a weight of the combination of the plurality of keywords in association with a combination of the plurality of keywords indicating the content of the image, and the weight calculation unit corresponds to the image extracted by the image extraction unit. The combination weight storage unit stores a combination of a plurality of keywords stored in the image storage unit in association with a keyword combination including the same keyword as the plurality of keywords extracted by the image keyword extraction unit. When the weights of combinations of a plurality of keywords are summed for each image extracted by the image extraction unit to further calculate a total value, and the image selection unit has the same total keyword weights calculated by the weight calculation unit In addition, the image extraction unit extracts an image having a larger total weight of combinations of a plurality of keywords calculated by the weight calculation unit. It may be selected from the image.

  Then, a keyword generation unit that generates a keyword indicating the content of the image based on the supplementary information of the image stored in the image storage unit, and the keyword generated by the keyword generation unit are stored in the image storage unit in association with the image. You may further provide the keyword provision part to be made. Further, an object extraction unit that extracts an object included in the image stored in the image storage unit, and an object storage unit that associates the object name of the object extracted by the object extraction unit with the image as a keyword indicating the content of the image. And an object name assigning unit to be stored. In addition, in a plurality of images stored in the image storage unit, for each object having an object name that is a keyword indicating the content of the image, a frequency calculating unit that calculates the number of images including the object, and a frequency calculating unit A frequency weight determining unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object having a larger number of images calculated for each object may be further included.

  The image storage unit further includes a position weight determination unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object closer to the center position of the image stored in the image storage unit. Preferably, the image storage unit further includes a size weight determination unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object having a larger size with respect to the image stored in the image storage unit. Also good. The image output unit further includes a trimming unit that trims the image selected by the image selection unit so that the object corresponding to the keyword assigned to the image by the object name assigning unit is included, and the image output unit inputs the image trimmed by the trimming unit. It may be output together with the written text.

  A sentence storage unit that stores a plurality of sentences in advance; and a sentence keyword weight storage unit that stores weights of the keywords in association with keywords included in the sentences stored in the sentence storage unit; The weight calculation unit is a keyword stored in the image storage unit in association with the image extracted by the image extraction unit, and is associated with the same keyword as the keyword extracted by the image keyword extraction unit. Further calculates the total value of the keyword weights stored in the image keyword, and when the total value of the keyword weights calculated by the weight calculation unit is the same, the text keyword weight storage unit calculated by the weight calculation unit An image having a larger total weight value of stored keywords may be selected from images extracted by the image extraction unit.

  And the text storage unit may store a plurality of texts created by the user in the past. In addition, for a plurality of sentences stored in the sentence storage unit, for each keyword included in the sentence stored in the sentence storage unit, a sentence keyword frequency calculation unit that calculates the number of keywords, and a sentence keyword frequency calculation unit May further include a sentence keyword weight determination unit that associates a keyword with a larger number of keywords calculated for each keyword and stores a larger weight in the sentence keyword weight storage unit. And an expression form weight storage unit that stores character typeface weights or character size weights in association with character typefaces or character sizes, and the sentence keyword weight determination unit is stored in the sentence storage unit. When the weight stored in the expression mode weight storage unit is larger in association with the typeface or character size of a predetermined keyword included in a sentence, the sentence keyword weight storage is performed by associating a larger weight with the predetermined keyword. You may make it store in a part.

  According to the second aspect of the present invention, there is provided an image processing method in which an image storage stage for storing an image in an image storage unit in association with a keyword indicating the content of the image is stored in the image storage unit. An image keyword weight storage stage for storing the keyword weight in the image keyword weight storage section in association with the keyword; an image keyword extraction stage for extracting a keyword stored in the image storage section from a newly input sentence; An image extraction stage for extracting an image stored in the image storage section in association with at least one keyword extracted in the image keyword extraction stage, and an image storage section in association with the image extracted in the image extraction stage Image keyword weight storage unit associated with the same keyword as the keyword extracted in the image keyword extraction stage. A weight calculation stage that calculates the total value by adding the weights of the stored keywords for each image extracted by the image extraction stage, and a predetermined number from the image that has a large total weight value calculated by the weight calculation stage Or an image selection stage in which the image extraction stage selects an image in which the total weight calculated by the weight calculation stage is larger than a predetermined value.

  According to a third aspect of the present invention, there is provided an image processing program for an image processing apparatus that extracts an image corresponding to a keyword included in an input sentence, the image processing apparatus being a keyword indicating the content of the image. An image storage unit that stores an image in association with a keyword, an image keyword weight storage unit that stores a weight of the keyword in association with a keyword stored in the image storage unit, and an image storage unit from a newly input sentence An image keyword extraction unit that extracts a keyword stored in the image storage unit, an image extraction unit that extracts an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit, and an image extraction unit Is a keyword stored in the image storage unit in association with the extracted image and has the same keyword as the keyword extracted by the image keyword extraction unit. The weight of the keyword stored in the image keyword weight storage unit in association with the word is summed for each image extracted by the image extraction unit, and the weight calculation unit calculates the total value. As an image selection unit that selects a predetermined number of images from an image with a large total value or an image in which the total value of weights calculated by the weight calculation unit is greater than a predetermined value from the images extracted by the image extraction unit Make it work.

  Further, in the fourth aspect of the present invention, the image storage unit that stores the image in association with the keyword indicating the content of the image, and the combination of a plurality of keywords stored in the image storage unit, A combination weight storage unit that stores weights of combinations of a plurality of keywords, an image keyword extraction unit that extracts a plurality of keywords stored in the image storage unit from newly input text, and an image keyword extraction unit A combination of an image extraction unit that extracts an image stored in the image storage unit in association with at least one keyword, and a plurality of keywords that are associated with the image extracted by the image extraction unit, and stores combination weights The image extraction unit extracts the weights of the keyword combinations that are the same as the keyword combinations stored in A weight calculating unit that calculates the total value of the weights for each image, and a predetermined number of images from the images that have a large total weight value of the combination of a plurality of keywords calculated by the weight calculating unit, or a weight calculating unit And an image selection unit that selects an image in which the total weight of the combination of the plurality of keywords calculated is greater than a predetermined value from the images extracted by the image extraction unit.

  According to a fifth aspect of the present invention, there is provided an image processing method in which an image storage stage storing an image in an image storage unit in association with a keyword indicating the content of the image is stored in the image storage unit. A combination weight storage stage for storing the weights of the combination of the plurality of keywords in the combination weight storage unit in association with the combination of the plurality of keywords, and a plurality of keywords stored in the image storage unit from newly input sentences Corresponding to at least one keyword extracted in the image keyword extraction stage, an image extraction stage that extracts an image stored in the image storage unit, and an image extracted in the image extraction stage A combination of a plurality of keywords, and a combination of a plurality of keywords stored in the combination weight storage unit The weight of the combination of a plurality of keywords identical to each other is calculated for each image extracted by the image extraction stage to calculate the total weight, and the weight of the combination of keywords calculated by the weight calculation stage An image in which the image extraction stage has extracted a predetermined number of images from an image having a large total value or an image in which the total weight of combinations of a plurality of keywords calculated by the weight calculation stage is greater than a predetermined value And an image selection step of selecting from.

  According to a sixth aspect of the present invention, there is provided an image processing program for an image processing apparatus for extracting an image corresponding to a keyword included in an input sentence, the image processing apparatus being a keyword indicating the content of the image. An image storage unit that stores images in association with each other, a combination weight storage unit that stores weights of combinations of the plurality of keywords in association with combinations of a plurality of keywords stored in the image storage unit, and a new An image keyword extraction unit that extracts a plurality of keywords stored in the image storage unit from the input text, and an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit A combination of an image extraction unit to be extracted and a plurality of keywords associated with the image extracted by the image extraction unit. A weight calculation unit that calculates a total weight value by adding the weights of a plurality of keyword combinations that are the same as a plurality of keyword combinations stored in the weight storage unit for each image extracted by the image extraction unit; A predetermined number of images from an image having a large total weight of a combination of a plurality of keywords calculated by the calculation unit, or a predetermined value of a total value of the weights of a plurality of keyword combinations calculated by the weight calculation unit A larger image is caused to function as an image selection unit that selects an image extracted by the image extraction unit.

  In the seventh embodiment of the present invention, a sentence storage unit that stores a plurality of sentences in advance, and a weight of the keyword is stored in association with a keyword included in the sentence stored in the sentence storage unit. A text keyword weight storage unit, an image storage unit that stores an image in association with a keyword indicating the content of the image, and an image keyword extraction unit that extracts a keyword stored in the image storage unit from newly input text An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit, and an image extraction unit for each at least one image extracted by the image extraction unit. Corresponds to the keyword stored in the image storage unit in association with the extracted image and the same keyword as the keyword extracted by the image keyword extraction unit A weight calculation unit for calculating the total value of the keyword weights stored in the text keyword weight storage unit, and a predetermined number of images from the images having a large total weight value calculated by the weight calculation unit, or weights And an image selection unit that selects an image whose total weight calculated by the calculation unit is larger than a predetermined value from the images extracted by the image extraction unit.

  According to an eighth aspect of the present invention, there is provided an image processing method in which the weight of the keyword is associated with a keyword included in a plurality of sentences stored in advance in the sentence storage unit in the sentence keyword weight storage unit. The sentence keyword weight storage stage to store, the image storage stage to store the image in the image storage section in association with the keyword indicating the content of the image, and the keyword stored in the image storage section from the newly input sentence An image keyword extraction stage, an image extraction stage for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction stage, and at least one image extracted by the image extraction stage A keyword stored in the image storage unit in association with the image extracted in the image extraction step, the image keyword extraction step A weight calculation stage for calculating a total value of keyword weights stored in the sentence keyword weight storage unit in association with the same keyword as the extracted keyword, and an image having a large total weight value calculated by the weight calculation stage in advance. An image selection step of selecting, from the images extracted by the image extraction step, a predetermined number of images or an image in which the total weight calculated by the weight calculation step is larger than a predetermined value.

  According to a ninth aspect of the present invention, there is provided an image processing program for an image processing apparatus for extracting an image corresponding to a keyword included in an input sentence, wherein the image processing apparatus stores a plurality of sentences in advance. A text storage unit, a text keyword weight storage unit that stores the weight of the keyword in association with a keyword included in the text stored in the text storage unit, and an image that is associated with a keyword that indicates the content of the image An image storage unit for storing image, an image keyword extraction unit for extracting a keyword stored in the image storage unit from newly input text, and an image storage associated with at least one keyword extracted by the image keyword extraction unit An image extracting unit that extracts an image stored in the image extracting unit, and an image extracted by the image extracting unit for each at least one image extracted by the image extracting unit. Corresponding keywords stored in the image storage unit, the total keyword weights stored in the text keyword weight storage unit are calculated in association with the same keywords extracted by the image keyword extraction unit A predetermined number of images from an image having a large total weight value calculated by the weight calculating unit, or an image in which the total weight value calculated by the weight calculating unit is greater than a predetermined value. The image extraction unit functions as an image selection unit that selects from the extracted images.

  The above summary of the invention does not enumerate all the necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

  According to the present invention, the image processing apparatus extracts a plurality of images associated with a keyword included in a sentence input by a user. Then, the image processing apparatus calculates the total value of the weights of the keywords that are included in the sentence and are associated with the image. Subsequently, since the image processing apparatus can select an image having the largest keyword weight total value, the user can simply input the sentence, and the image processing apparatus is suitable for the keyword included in the sentence. The recalled image can be automatically selected.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the scope of claims, and all combinations of features described in the embodiments are included. It is not necessarily essential for the solution of the invention.

  FIG. 1 shows an example of a functional configuration of an image processing apparatus 10 according to an embodiment of the present invention. The image processing apparatus 10 includes a text input unit 100, an image input unit 102, a text storage unit 104, an image storage unit 110, an image keyword weight storage unit 120, a combination weight storage unit 122, a text keyword weight storage unit 124, and a text keyword frequency calculation. Unit 126, expression mode weight storage unit 128, image keyword extraction unit 130, image extraction unit 140, weight calculation unit 150, image weight determination unit 160, image selection unit 170, trimming unit 172, image output unit 174, sentence keyword weight determination Unit 180, keyword generation unit 190, keyword assignment unit 192, object extraction unit 194, and object name assignment unit 196.

  The image processing apparatus 10 according to the present embodiment extracts a plurality of images associated with keywords included in a sentence input by a user. Then, the image processing apparatus 10 calculates a total value of the weights of the keywords that are included in the sentence and are associated with the image. Subsequently, the image processing apparatus 10 automatically selects an image that is reconciled with the keyword included in the sentence input by the user and selected by the user by selecting an image having the largest total keyword weight. The purpose is to do.

  The text input unit 100 inputs text to the image processing apparatus 10. The text input unit 100 supplies the input text to the image keyword extraction unit 130 and the image output unit 174. The text input unit 100 supplies the input text to the image output unit 174. The image input unit 102 inputs an image to the image processing apparatus 10. Then, the image input unit 102 supplies the input image to the image storage unit 110. The image storage unit 110 stores an image in association with a keyword indicating the content of the image. The image storage unit 110 may store a plurality of keywords in association with each image. The keyword indicating the content of the image is determined based on a keyword indicating the object name of the object included in the image, a keyword indicating the background included in the image, and an imaging condition attached to the image at the time of image capturing. It may be a keyword or the like. The image storage unit 110 supplies the stored image to the image extraction unit 140, the keyword generation unit 190, the object extraction unit 194, and the image weight determination unit 160. Further, the image storage unit 110 supplies the stored keywords to the image keyword weight storage unit 120 and the combination weight storage unit 122.

  The keyword generation unit 190 generates a keyword indicating the content of the image based on the supplementary information of the image stored in the image storage unit 110. Specifically, the keyword generation unit 190 extracts auxiliary information attached to the image stored in the image storage unit 110. The incidental information includes, for example, the image capturing date and time, the image capturing location, the image capturing mode (such as the close-up mode, the portrait mode, and the night view mode) and the image capturing conditions (such as brightness, saturation, and focusing distance) of the image. Good. For example, for an image captured in the night view mode, the keyword generation unit 190 generates “night” as a keyword for the image. In addition, the keyword generation unit 190 may generate the year, month, or day when the image was captured from the date and time when the image was captured as a keyword. Further, the keyword generation unit 190 may generate the season in which the image is captured as a keyword from the image capturing date and time. For example, the keyword generation unit 190 captures “spring” for images captured from March to April, “summer” for images captured from June to August, and images from September to November. “Autumn” may be generated as a keyword for images captured in the image, and “Winter” may be generated as a keyword for images captured from December to February.

  The keyword generation unit 190 may generate keywords such as “morning”, “daytime”, “evening”, and “night” based on the imaging time. Further, the keyword generation unit 190 may generate, as keywords, the name of the area where the image is captured, the name of the place, the name of the city, and the like based on the imaging location. Then, based on the imaging conditions, the keyword generation unit 190 may generate “bright” or the like as a keyword when the brightness is high, or “bright” or “flashy” or the like as a keyword when the saturation is high. Good. The keyword generation unit 190 supplies the generated keyword to the keyword assignment unit 192. The keyword assigning unit 192 stores the keyword generated by the keyword generating unit 190 in the image storage unit 110 in association with the image. The keyword assigning unit 192 may associate a plurality of keywords with the image and store the plurality of keywords in the image storage unit 110 in association with the image.

  The object extraction unit 194 extracts an object included in the image stored in the image storage unit 110. For example, the object extraction unit 194 extracts an object using image processing such as edge detection. Further, the object extraction unit 194 may extract the object and the background by dividing into an area where the spatial frequency component of the image is higher than a predetermined frequency and an area excluding the area. The object extracting unit 194 supplies the extracted object to the object name assigning unit 196. The object name assigning unit 196 associates the object name of the object extracted by the object extracting unit 194 with the image as a keyword indicating the content of the image. Then, the object name assigning unit 196 stores the keyword associated with the image in the image storage unit 110.

  For example, the object name assigning unit 196 may have an object template database that stores object names in association with template images. The object name assigning unit 196 matches the object received from the object extracting unit 194 with the template image stored in the object template database, and associates the object name stored in the object template database with the matched template image. The keyword may be associated with the image as a keyword indicating the content of the image. For example, when it is found by matching with the template image that the object included in the image is a cat, the object name assigning unit 196 may associate “cat” with the image as a keyword indicating the content of the image. The object name assigning unit 196 may store the keyword input by the user in the image storage unit 110 in association with the image.

  The sentence storage unit 104 stores a plurality of sentences in advance. Specifically, a plurality of sentences created by the user in the past are stored. The text storage unit 104 supplies the text to the text keyword frequency calculation unit 126 and the expression mode weight storage unit 128. The sentence keyword frequency calculation unit 126 calculates the number of keywords for each keyword included in the plurality of sentences received from the sentence storage unit 104. For example, when the keyword is “A”, the sentence keyword frequency calculation unit 126 calculates the number of keywords “A” included in the plurality of sentences. The text keyword frequency calculation unit 126 supplies the calculated keyword number for each keyword to the text keyword weight determination unit 180. In addition, the keyword when the sentence keyword frequency calculation unit 126 calculates may be extracted by using natural language processing such as morphological analysis, and the number of times the keyword appears in the sentence may be calculated for each extracted keyword. .

  The expression form weight storage unit 128 stores the weight of the character font or the character size in association with the character font or character size. Specifically, the expression mode weight storage unit 128 stores a weight associated with each typeface such as Mincho, Gothic, and line typeface. In addition, the expression mode weight storage unit 128 stores the weight associated with each character expression mode, such as the size of the character, whether it is italic, and whether it is bold. When the user changes the typeface or size of the character with respect to the character, the user often wants to display the character more emphasized than other characters. Therefore, for example, when the character size is large, the expression mode weight storage unit 128 may store a larger weight than the character size with a smaller character size. The expression mode weight storage unit 128 supplies the weight stored in association with the font or size of a predetermined keyword to the text keyword weight determination unit 180.

  The sentence keyword weight determination unit 180 associates a larger weight with a keyword having a larger number of keywords for each keyword received from the sentence keyword frequency calculation unit 126. Also, the sentence keyword weight determination unit 180 has a larger weight stored in the expression mode weight storage unit 128 in association with the font or character size of a predetermined keyword included in the sentence stored in the sentence storage unit 104. In some cases, a larger weight is associated with a predetermined keyword. For example, the sentence keyword weight determination unit 180 adds the keyword weight calculated by the sentence keyword frequency calculation unit 126 and the weight stored in the expression form weight storage unit 128 in association with the typeface and size of the keyword. Thus, the weight of the keyword may be calculated. The text keyword weight determination unit 180 calculates the weight of the keyword by multiplying the keyword weight by the weight stored in the expression form weight storage unit 128 in association with the typeface of the keyword. Also good. The sentence keyword weight determination unit 180 supplies the weight determined in association with the keyword to the sentence keyword weight storage unit 124. The sentence keyword weight storage unit 124 stores the keyword weight in association with the keyword included in the sentence stored in the sentence storage unit 104. Specifically, the sentence keyword weight storage unit 124 receives and stores the weight for each keyword determined by the sentence keyword weight determination unit 180. The text keyword weight storage unit 124 supplies the image selection unit 170 with a weight stored in association with the keyword.

  Accordingly, the text keyword weight determination unit 180 can associate a greater weight with a keyword that is frequently input by the user than a keyword that is less input by the user. Reference numeral 10 refers to the weight associated with the keyword included in the sentence, so that it is possible to accurately grasp the keyword that is highly interested by the user and often recalled by the user. Further, the sentence keyword weight determination unit 180 can weight the predetermined keyword based on the typeface or size of the character set by the user as the predetermined keyword, and accordingly, according to the interest when the user is inputting the sentence. An appropriate weight can be associated with a keyword included in the sentence.

  The image keyword weight storage unit 120 stores keyword weights in association with the keywords stored in the image storage unit 110. Further, the image keyword weight storage unit 120 may store keyword weights based on the control of the image weight determination unit 160. The image keyword weight storage unit 120 supplies the keyword weight to the weight calculation unit 150. The combination weight storage unit 122 stores the weight of the combination of the plurality of keywords in association with the combination of the plurality of keywords indicating the content of the image. The combination weight storage unit 122 supplies the weights of combinations of a plurality of keywords to the weight calculation unit 150. The image weight determination unit 160 calculates a weight for each of a plurality of image keywords received from the image storage unit 110, and stores the calculated weight in the image keyword weight storage unit 120 in association with the image keyword. Specifically, the image weight determination unit 160 includes a frequency calculation unit 162, a frequency weight determination unit 164, a position weight determination unit 166, and a size weight determination unit 168.

  The frequency calculation unit 162 calculates the number of images including the object for each object having an object name that is a keyword indicating the content of the image in the plurality of images stored in the image storage unit 110. The frequency calculation unit 162 also associates the keywords associated with each of the plurality of images stored in the image storage unit 110 with the plurality of images stored in the image storage unit 110. The number of keywords may be calculated for each keyword. The frequency calculation unit 162 supplies the frequency weight determination unit 164 with the number of images including the calculated object for each object and the number of the keyword for each keyword associated with the image included in a plurality of images. To do. The frequency weight determination unit 164 stores a larger weight in the image keyword weight storage unit 120 in association with a keyword that is an object name having a larger number of images calculated for each object by the frequency calculation unit 162. Further, the frequency weight determining unit 164 stores a larger weight in the image keyword weight storage unit 120 in association with a keyword having a larger number of keywords included in a plurality of images calculated by the frequency calculating unit 162 for each keyword. You may let them.

  As a result, when the user captures more images of the predetermined object, the frequency weight determination unit 164 can apply a higher weight to the image including the object, and thus the user is highly interested. An image including an object can be given a higher weight than an image including an object that is not of interest to the user.

  The position weight determination unit 166 stores a larger weight in the image keyword weight storage unit 120 in association with a keyword that is an object name of an object closer to the center position of the image stored in the image storage unit 110. For example, the position weight determination unit 166 calculates the position occupied by the object included in the image with respect to the image. Subsequently, the position weight determination unit 166 calculates the distance from the center position of the image to the object. Then, the position weight determination unit 166 associates a greater weight with the keyword that is the object name of the object in association with the distance from the center position of the image to the object, when the distance is closer. Here, when the user captures an image, an object that the user desires to capture is often included in an area close to the center position of the image. Therefore, since the image processing apparatus 10 according to the present embodiment includes the position weight determination unit 166, a larger weight is applied to an image including an object with high user interest than an image including an object with low user interest. Can be weighted.

  The size weight determination unit 168 causes the image keyword weight storage unit 120 to store a larger weight in association with the keyword that is the object name of the object having a larger size with respect to the image stored in the image storage unit 110. For example, the size weight determination unit 168 calculates the area of the imaging region of the image and the area of the object included in the image. Subsequently, the size weight determination unit 168 calculates the ratio of the area of the object to the image. Then, the size weight determination unit 168 may associate a larger weight with the keyword that is the object name of the object in association with the proportion of the area of the object in the image when the proportion is larger.

  The image keyword extraction unit 130 extracts a keyword stored in the image storage unit 110 from a sentence newly input by the user. Specifically, the image keyword extraction unit 130 extracts keywords stored in the image storage unit 110 from the text received from the text input unit 100 using a processing method such as natural language processing. The image keyword extraction unit 130 supplies the extracted keyword to the image extraction unit 140. The image extraction unit 140 extracts an image stored in the image storage unit 110 in association with at least one keyword extracted by the image keyword extraction unit 130. Specifically, the image extraction unit 140 extracts, from the image storage unit 110, an image that is associated with the same keyword as at least one of the keywords included in the input sentence. The image extraction unit 140 supplies the extracted image to the image selection unit 170. In addition, the image extraction unit 140 supplies the keyword stored in the image storage unit 110 in association with the extracted image to the weight calculation unit 150.

  The weight calculation unit 150 is a keyword stored in the image storage unit 110 in association with the image received from the image extraction unit 140 and is associated with the same keyword as the keyword extracted by the image keyword extraction unit 130. The keyword weights stored in the keyword weight storage unit 120 are extracted. Then, the weight calculation unit 150 adds the extracted keyword weights for each image to calculate a total value. In addition, the weight calculation unit 150 sets a plurality of keywords stored in the image storage unit 110 in association with the image received from the image extraction unit 140 and the same keyword as the keyword extracted by the image keyword extraction unit 130. The combination weights stored in the combination weight storage unit 122 are extracted in association with each other. Then, the weight calculation unit 150 calculates the total value by summing the weights of the extracted keyword combinations for each image. Further, the weight calculation unit 150 associates the keyword stored in the image storage unit 110 with the image received from the image extraction unit 140 and the same keyword as the keyword extracted by the image keyword extraction unit 130. Thus, the total value of the keyword weights stored in the sentence keyword weight storage unit 124 may be calculated. The weight calculation unit 150 supplies the calculated weight for each image to the image selection unit 170.

  The image selection unit 170 selects a predetermined number of images from the images received from the image extraction unit 140 from images having a large total weight value received from the weight calculation unit 150. In addition, the image selection unit 170 may select, from the images received from the image extraction unit 140, an image whose total weight value received from the weight calculation unit 150 is larger than a predetermined value. Further, the image selection unit 170 may select an image with the maximum total weight received from the weight calculation unit 150 from the images received from the image extraction unit 140. Then, when there are a plurality of images having the same keyword weight total value received from the weight calculation unit 150, the image selection unit 170 obtains the total weight value of the combination of the plurality of keywords calculated by the weight calculation unit 150. A large image may be selected from the images received from the image extraction unit 140. Further, when there are a plurality of images having the same keyword weight total value received from the weight calculation unit 150, the image selection unit 170 stores the keyword stored in the text keyword weight storage unit 124 calculated by the weight calculation unit 150. An image having a larger total weight may be selected from the images received from the image extraction unit 140. The image selection unit 170 supplies the selected image to the trimming unit 172 and the image output unit 174.

  The trimming unit 172 trims the image received from the image selection unit 170 so that the object corresponding to the keyword assigned to the image by the object name assigning unit 196 is included. For example, the trimming unit 172 may perform trimming using position information of the object attached to the image in association with the object included in the image. In addition, the trimming unit 172 may trim the object included in the image so as to include an object corresponding to the keyword added to the image by the object name adding unit 196 by image processing such as pattern matching. Further, the trimming unit 172 may enlarge and reduce the object included in the image. The trimming unit 172 supplies the trimmed image to the image output unit 174.

  The image output unit 174 outputs the image selected by the image selection unit 170 together with the text received from the text input unit 100. The image output unit 174 may output the trimmed image received from the trimming unit 172 together with the input text. For example, the image output unit 174 outputs an image together with text on a monitor or the like. Specifically, the image output unit 174 records a sentence or the like together with the URL of the web page on a website on the Internet. When the user inputs a sentence on a blog or the like, the image output unit 174 selects a keyword included in the sentence inputted by the user. The image selected in the above description is output together with the text.

  According to the image processing apparatus 10 according to the present embodiment, the keywords included in the input sentence, the total weight of the keyword associated with the image and the combination of the plurality of keywords associated with the image. Calculate the value. Then, when there are a plurality of images having the largest keyword weight total value or images having the same keyword weight total value, the image processing apparatus 10 has the largest total weight value of the combination of the plurality of keywords. An image can be selected. As a result, the image processing apparatus 10 can automatically select an image suitable for the content of the text input by the user and can be displayed together with the text. .

  FIG. 2 shows an example of the data structure of the image storage unit 110 according to this embodiment. The image storage unit 110 stores an image in association with a keyword indicating the content of the image. The image stored in the image storage unit 110 may be an image captured by the user in the past, an image received from a network such as the Internet, an image created by the user, or the like. Then, the image storage unit 110 stores an image in association with a keyword indicating the content of at least one image received from the keyword assigning unit 192 and the object name assigning unit 196. Furthermore, the image storage unit 110 may store a keyword indicating the content of the image input by the user in association with the image in association with the image. Note that there may be a plurality of keywords stored in the image storage unit 110 in association with images.

  Specifically, the image storage unit 110 stores, for each image, at least one keyword indicating the content of the image in association with the image. For example, the image storage unit 110 stores the keyword A, the keyword B, and the keyword D in association with the image 300. Further, keyword B, keyword C, and keyword D are stored in association with image 310. Further, the image storage unit 110 stores the keyword A, the keyword C, and the keyword F in the image 320, and the keyword 330, the keyword G, the keyword I, the keyword K, and the like in association with each other.

  FIG. 3 shows an example of the data structure of the image keyword weight storage unit 120 according to the present embodiment. The image keyword weight storage unit 120 stores the keyword weight in association with the keyword stored in the image storage unit 110. Specifically, the image keyword weight storage unit 120 stores the keyword weight received from the image weight determination unit 160 in association with the keyword. For example, the image keyword weight storage unit 120 stores the keyword A in association with the weight “3”. Further, the image keyword weight storage unit 120 stores the keyword “B” in association with the keyword “B”. The image keyword weight storage unit 120 stores the keyword C in association with the weight “1”. Further, the image keyword weight storage unit 120 stores the weight “4” in association with the keyword D.

  The weight calculation unit 150 extracts the weight of the keyword associated with the image from the image keyword weight storage unit 120. Then, the weight calculator 150 calculates the total value of the keyword weights associated with the image. For example, consider a case where the keywords extracted from the text input by the image keyword extraction unit 130 from the text input unit 100 are the keyword “A” and the keyword “D”. In such a case, the image extraction unit 140 extracts an image associated with at least one of the keyword “A” or the keyword “D” from the image storage unit 110. That is, the image extraction unit 140 retrieves the image 300, the image 310, and the image 320 associated with at least one of the keyword “A” and the keyword “D” from the image storage unit 110 in the above description of FIG. Extract.

  Then, the weight calculation unit 150 calculates the total value of the weight of the keyword “A” and the weight of the keyword “D” associated with each of the image 300, the image 310, and the image 320 for each image. For example, the keyword “A”, the keyword “B”, and the keyword “D” are associated with the image 300. In this case, it is assumed that the same keyword as the keyword extracted by the image keyword extraction unit 130 among the keywords associated with the image 300 is the keyword “A” and the keyword “D”. In this case, the weight calculation unit 150 calculates the weight of the image 300 as “7”, which is the sum of the weight of the keyword “A” and the weight of the keyword “D”. Similarly, the weight calculation unit 150 calculates the total weight of the image 310 as “2”. Further, the weight calculation unit 150 calculates the total weight of the image 320 as “3”.

  Subsequently, the image selection unit 170 compares the total weight value calculated by the weight calculation unit 150 for each image, and selects the image having the maximum weight value. In the description of FIG. 2 and FIG. 3 described above, the image selection unit 170 selects the image 300 having the maximum weight value. Then, the image output unit 174 displays the image 300 selected by the image selection unit 170 together with the text input by the user. Note that the image selection unit 170 may select a predetermined number of images from images having a large total weight calculated by the weight calculation unit 150. Specifically, the image selection unit 170 arranges a plurality of images in association with the descending order of the total weight value calculated by the weight calculation unit 150, and sets a predetermined number of images with the maximum total weight value. You may select from the image in order. The image selection unit 170 may select an image in which the total weight calculated by the weight calculation unit 150 is greater than a predetermined value. In such a case, there may be a plurality of images selected by the image selection unit 170.

  According to the image processing apparatus 10 according to the present embodiment, an image associated with the same keyword as at least one of the keywords included in the text input by the user is extracted, and the keyword associated with the image is extracted. A weight is calculated for each image. Since the keyword weight total value calculated for each image can be compared and the image can be automatically selected, the image recalled by the user can be appropriately selected and displayed together with the text input by the user. For example, according to the image processing apparatus 10 according to the present embodiment, an image recalled by the user is automatically selected based on the keyword weight included in the text input by the user on a blog or the like, and the text is combined with the text. Can be displayed.

  FIG. 4 shows an example of the data structure of the combination weight storage unit 122 according to this embodiment. The combination weight storage unit 122 stores the weight of the combination of the plurality of keywords in association with the combination of the plurality of keywords indicating the content of the image. Specifically, the combination weight storage unit 122 stores weights based on associations between keywords of a plurality of keywords associated with images. More specifically, among the plurality of images stored in the image storage unit 110, when the number of combinations of a plurality of keywords associated with the image is larger, the combination of the plurality of keywords is larger. Store weights in association with each other. For example, consider a case where a user frequently takes images of winter mountains and stores a plurality of images of the taken winter mountains in the image storage unit 110. In such a case, the image storage unit 110 has more winter mountain images associated with the keywords “winter” and “mountain” than images associated with keywords other than the keywords “winter” and “mountain”. Since it is stored, the relationship between the keywords “winter” and “mountain” is high, and a combination of keywords of “winter” and “mountain” may be associated with a larger weight and stored.

  For example, the combination weight storage unit 122 stores the weight “4” in association with the combination of the keyword “A” and the keyword “B”. The combination weight storage unit 122 stores the weight “3” in association with the combination of the keyword “A” and the keyword “D”. Similarly, the combination weight storage unit 122 assigns a weight “2” to the combination of the keyword “A” and the keyword “F”, a weight “5” to the combination of the keyword “B” and the keyword “D”, and the keyword “D”. ”And the keyword“ F ”are stored in association with the weight“ 1 ”. Note that the combination weight storage unit 122 is not limited to the combination of two keywords, and may store weights in association with a combination of at least two or more keywords. The image processing apparatus 10 may further include a combination weight determining unit that determines combination weights between a plurality of keywords. The combination weight determining unit calculates the number of combinations of a plurality of keywords associated with the image, and the combination having a larger number of combinations of the plurality of keywords is compared to a combination having a smaller number of combinations of the plurality of keywords. Larger weights may be associated.

  FIG. 5 shows an example of the flow of processing in the image processing apparatus 10 according to the present embodiment. First, the image keyword extraction unit 130 extracts keywords from the text newly received by the user, received from the text input unit 100 (S1000). The image keyword extraction unit 130 supplies the extracted keyword to the image extraction unit 140. The image extraction unit 140 extracts an image corresponding to the keyword received from the image keyword extraction unit 130 from the image stored in the image storage unit 110 (S1010). Then, the image extraction unit 140 supplies the weight of the keyword associated with each extracted image to the weight calculation unit 150. The image extraction unit 140 supplies the image extracted from the image storage unit 110 to the image selection unit 170.

  Subsequently, the weight calculation unit 150 extracts the weight stored in the image keyword weight storage unit 120 in association with the keyword received from the image extraction unit 140, and calculates the total value of the keyword weight for each image. (S1020). Further, the weight calculation unit 150 extracts the weight stored in the combination weight storage unit 122 in association with the combination of a plurality of keywords associated with the image, and calculates the weight of the combination of the plurality of keywords for each image. The total value is also calculated (S1030). The weight calculation unit 150 supplies the calculated total weight value of the keyword and the total weight value of the combination of the plurality of keywords to the image selection unit 170.

  The image selection unit 170 determines whether there is only one image having the largest keyword weight total value received from the weight calculation unit 150 (S1040). When only one image having the largest keyword weight total value is extracted from the images extracted by the image extraction unit 140 (S1040: Yes), the image selection unit 170 has the largest keyword weight total value. An image is selected (S1050). On the other hand, if the image extracted by the image extraction unit 140 is not the only image with the largest total keyword weight (S1040: No), the image selection unit 170 may combine a plurality of keywords. An image having a larger total value is selected (S1060).

  Here, the image selection unit 170 may select only an image having the maximum total keyword weight received from the weight calculation unit 150. Further, the image selection unit 170 may select only the image having the maximum total weight of the combination of the plurality of keywords received from the weight calculation unit 150. Further, the image selection unit 170 may first select an image having the maximum total weight of the plurality of keywords received from the weight calculation unit 150. When a plurality of images having the maximum total weight of the combination of a plurality of keywords are selected, the image selection unit 170 may select an image having the maximum total keyword weight.

  The image selection unit 170 supplies the selected image to the image output unit 174. The image output unit 174 outputs the image selected by the image selection unit 170 together with the text input by the text input unit 100 (S1070). For example, the image output unit 174 displays the image washed by the image selection unit 170 together with the text on a monitor or the like. The image output unit 174 may display an image on the blog together with the text input by the user when the user inputs the text on the blog or the like. The image processing apparatus according to claims 17 and 20 is an example of the image processing apparatus described with reference to FIGS.

  According to the image processing apparatus 10 according to the present embodiment, the keywords included in the input sentence, including the total value of the keyword weights associated with the image and the plurality of keywords associated with the image. The total value of the combination weights can be calculated. Then, the image processing apparatus 10 converts the total weight of the keywords and the total weight of the combinations of the keywords from a plurality of images associated with the same keyword as the keyword included in the text input by the user. Images can be automatically selected based on. As a result, the image processing apparatus 10 is suitable for the content of the text input by the user and can automatically select an image suitable for the image recalled by the user, so that the user's image can be obtained without making the user feel uncomfortable. The same image can be displayed along with the text.

  FIG. 6 shows an example of the hardware configuration of the image processing apparatus 10 according to the present embodiment. The image processing apparatus 10 according to this embodiment includes a CPU peripheral unit including a CPU 1505, a RAM 1520, a graphic controller 1575, and a display device 1580 that are connected to each other by a host controller 1582, and a host controller 1582 by an input / output controller 1584. I / O unit having a communication interface 1530, a hard disk drive 1540, and a CD-ROM drive 1560 connected to the I / O controller, and a legacy input having a ROM 1510, a flexible disk drive 1550, and an I / O chip 1570 connected to the I / O controller 1584. And an output unit.

  The host controller 1582 connects the RAM 1520 to the CPU 1505 and the graphic controller 1575 that access the RAM 1520 at a high transfer rate. The CPU 1505 operates based on programs stored in the ROM 1510 and the RAM 1520 to control each unit. The graphic controller 1575 acquires image data generated by the CPU 1505 or the like on a frame buffer provided in the RAM 1520 and displays it on the display device 1580. Alternatively, the graphic controller 1575 may include a frame buffer that stores image data generated by the CPU 1505 or the like.

  The input / output controller 1584 connects the host controller 1582 to the communication interface 1530, the hard disk drive 1540, and the CD-ROM drive 1560, which are relatively high-speed input / output devices. The communication interface 1530 communicates with other devices via a network. The hard disk drive 1540 stores programs and data used by the CPU 1505 in the image processing apparatus 10. The CD-ROM drive 1560 reads a program or data from the CD-ROM 1595 and provides it to the hard disk drive 1540 via the RAM 1520.

  The input / output controller 1584 is connected to the ROM 1510, the flexible disk drive 1550, and the relatively low-speed input / output device of the input / output chip 1570. The ROM 1510 stores a boot program that the image processing apparatus 10 executes at startup, a program that depends on the hardware of the image processing apparatus 10, and the like. The flexible disk drive 1550 reads a program or data from the flexible disk 1590 and provides it to the hard disk drive 1540 via the RAM 1520. The input / output chip 1570 connects various input / output devices via a flexible disk drive 1550 such as a parallel port, a serial port, a keyboard port, and a mouse port.

  The image processing program provided to the hard disk drive 1540 via the RAM 1520 is stored in a recording medium such as the flexible disk 1590, the CD-ROM 1595, or an IC card and provided by the user. The image processing program is read from the recording medium, installed in the hard disk drive 1540 in the image processing apparatus 10 via the RAM 1520, and executed by the CPU 1505. The image processing program installed and executed in the image processing apparatus 10 works on the CPU 1505 or the like to cause the image processing apparatus 10 to be described with reference to FIGS. 1 to 5, the text input unit 100, the image input unit 102, and the text storage unit 104. , Image storage unit 110, image keyword weight storage unit 120, combination weight storage unit 122, sentence keyword weight storage unit 124, sentence keyword frequency calculation unit 126, expression mode weight storage unit 128, image keyword extraction unit 130, image extraction unit 140 , Weight calculation unit 150, image weight determination unit 160, image selection unit 170, trimming unit 172, image output unit 174, sentence keyword weight determination unit 180, keyword generation unit 190, keyword assignment unit 192, object extraction unit 194, object name Giving unit 196, frequency calculating unit 162, frequency See decision unit 164, to function as a position weight determining section 166 and the size weight determining unit 168,.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

2 is a block diagram showing a functional configuration of the image processing apparatus 10. FIG. 3 is a diagram illustrating a data structure of an image storage unit 110. FIG. 3 is a diagram illustrating a data structure of an image keyword weight storage unit 120. FIG. 6 is a diagram illustrating a data structure of a combination weight storage unit 122. FIG. 3 is a flowchart showing a flow of processing in the image processing apparatus 10. 2 is a block diagram illustrating a hardware configuration of the image processing apparatus 10. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image processing apparatus 100 Text input part 102 Image input part 104 Text storage part 110 Image storage part 120 Image keyword weight storage part 122 Combination weight storage part 124 Text keyword weight storage part 126 Text keyword frequency calculation part 128 Expression mode weight storage part 130 Image keyword extraction unit 140 Image extraction unit 150 Weight calculation unit 160 Image weight determination unit 162 Frequency calculation unit 164 Frequency weight determination unit 166 Position weight determination unit 168 Size weight determination unit 170 Image selection unit 172 Trimming unit 174 Image output unit 180 Text Keyword weight determination unit 190 Keyword generation unit 192 Keyword assignment unit 194 Object extraction unit 196 Object name assignment unit 300, 310, 320, 330 Image 1505 CPU
1510 ROM
1520 RAM
1530 Communication interface 1540 Hard disk drive 1550 Flexible disk drive 1560 CD-ROM drive 1570 Input / output chip 1575 Graphic controller 1580 Display device 1582 Host controller 1584 Input / output controller 1590 Flexible disk 1595 CD-ROM

Claims (22)

An image storage unit for storing an image in association with a keyword indicating the content of the image;
An image keyword weight storage unit that stores a weight of the keyword in association with a keyword stored in the image storage unit;
An image keyword extraction unit that extracts a keyword stored in the image storage unit from a newly input sentence;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A keyword stored in the image storage unit in association with the image extracted by the image extraction unit, the image keyword weight storage unit in association with the same keyword as the keyword extracted by the image keyword extraction unit. A weight calculation unit that calculates the total value by adding the weights of the stored keywords for each image extracted by the image extraction unit;
A predetermined number of images from an image having a large total weight value calculated by the weight calculation unit or an image in which the total weight value calculated by the weight calculation unit is larger than a predetermined value An image processing apparatus comprising: an image selection unit that selects an image extracted from the image. The image processing apparatus according to claim 1, further comprising an image output unit that outputs the image selected by the image selection unit together with the input text. The image processing apparatus according to claim 2, wherein the image storage unit stores images captured by the user in the past. A combination weight storage unit that stores a weight of the combination of the plurality of keywords in association with a combination of the plurality of keywords indicating the content of the image,
The weight calculation unit is a combination of a plurality of keywords stored in the image storage unit in association with the image extracted by the image extraction unit, and is the same as the plurality of keywords extracted by the image keyword extraction unit The total weight is further calculated by summing the weights of the combinations of a plurality of keywords stored in the combination weight storage unit in association with the keyword combinations consisting of keywords for each image extracted by the image extraction unit,
The image selection unit is configured to extract an image having a larger total weight value of a plurality of keyword combinations calculated by the weight calculation unit when the total weight value of the keywords calculated by the weight calculation unit is the same. The image processing device according to claim 2, wherein the image is selected from images extracted by the unit. A keyword generation unit for generating a keyword indicating the content of the image based on the incidental information of the image stored in the image storage unit;
The image processing apparatus according to claim 2, further comprising: a keyword adding unit that stores the keyword generated by the keyword generating unit in the image storage unit in association with an image. An object extraction unit for extracting an object included in the image stored in the image storage unit;
The image processing apparatus according to claim 2, further comprising: an object name adding unit that stores the object name of the object extracted by the object extracting unit in the image storage unit in association with the image as a keyword indicating the content of the image. A frequency calculation unit that calculates the number of images including the object for each object having an object name that is a keyword indicating the content of the image in the plurality of images stored in the image storage unit;
A frequency weight determining unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object having a larger number of images calculated for each object by the frequency calculation unit. Item 7. The image processing apparatus according to Item 6. A position weight determination unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object closer to the center position of the image stored in the image storage unit. 6. The image processing apparatus according to 6. A size weight determining unit that stores a larger weight in the image keyword weight storage unit in association with a keyword that is an object name of an object having a larger size with respect to an image stored in the image storage unit. Item 7. The image processing apparatus according to Item 6. A trimming unit for trimming the image selected by the image selection unit so as to include an object corresponding to the keyword given to the image by the object name giving unit;
The image processing apparatus according to claim 6, wherein the image output unit further includes outputting the image trimmed by the trimming unit together with the input text. A sentence storage unit storing a plurality of sentences in advance;
A text keyword weight storage unit that stores a weight of the keyword in association with a keyword included in the text stored in the text storage unit;
The weight calculation unit is a keyword stored in the image storage unit in association with the image extracted by the image extraction unit, and is associated with the same keyword as the keyword extracted by the image keyword extraction unit. Further calculating the total value of the keyword weights stored in the sentence keyword weight storage unit,
When the total value of the keyword weights calculated by the weight calculation unit is the same, the image selection unit has a keyword total value stored in the sentence keyword weight storage unit calculated by the weight calculation unit. The image processing apparatus according to claim 6, wherein a large image is selected from images extracted by the image extraction unit. The image processing apparatus according to claim 11, wherein the sentence storage unit stores a plurality of sentences created by a user in the past. In a plurality of sentences stored in the sentence storage unit, for each keyword included in the sentence stored in the sentence storage unit, a sentence keyword frequency calculation unit that calculates the number of the keywords;
The sentence keyword weight determination part which associates with the keyword with more keywords calculated by the sentence keyword frequency calculation part for each keyword and stores a larger weight in the sentence keyword weight storage part. The image processing apparatus described. An expression mode weight storage unit that stores the weight of the font of the character or the weight of the character size in association with the font of the character or the size of the character;
When the weight stored in the expression mode weight storage unit is larger in association with the font or character size of a predetermined keyword included in the text stored in the text storage unit, the text keyword weight determination unit The image processing apparatus according to claim 13, wherein a larger weight is associated with the predetermined keyword and stored in the sentence keyword weight storage unit. An image storage stage for storing the image in the image storage unit in association with a keyword indicating the content of the image;
An image keyword weight storage step of storing the weight of the keyword in the image keyword weight storage unit in association with the keyword stored in the image storage unit;
An image keyword extraction step for extracting a keyword stored in the image storage unit from a newly input sentence;
An image extraction step of extracting an image stored in the image storage unit in association with at least one keyword extracted in the image keyword extraction step;
The keyword stored in the image storage unit in association with the image extracted in the image extraction step, the image keyword weight storage unit in association with the same keyword as the keyword extracted in the image keyword extraction step. A weight calculating step of calculating a total value by adding weights of stored keywords for each image extracted by the image extracting step;
The image extracting step calculates a predetermined number of images from an image having a large total weight value calculated by the weight calculating step, or an image having a total weight value calculated by the weight calculating step larger than a predetermined value. An image processing method comprising: an image selection step of selecting from images extracted from the image. An image processing program for an image processing apparatus for extracting an image corresponding to a keyword included in an input sentence, the image processing apparatus comprising:
An image storage unit for storing an image in association with a keyword indicating the content of the image;
An image keyword weight storage unit that stores a weight of the keyword in association with a keyword stored in the image storage unit;
An image keyword extraction unit that extracts a keyword stored in the image storage unit from a newly input sentence;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A keyword stored in the image storage unit in association with the image extracted by the image extraction unit, the image keyword weight storage unit in association with the same keyword as the keyword extracted by the image keyword extraction unit. A weight calculation unit that calculates the total value by adding the weights of the stored keywords for each image extracted by the image extraction unit;
A predetermined number of images from an image having a large total weight value calculated by the weight calculation unit or an image in which the total weight value calculated by the weight calculation unit is larger than a predetermined value Is an image processing program that functions as an image selection unit that selects from images extracted by. An image storage unit for storing an image in association with a keyword indicating the content of the image;
A combination weight storage unit that stores weights of combinations of the plurality of keywords in association with combinations of the plurality of keywords stored in the image storage unit;
An image keyword extraction unit that extracts a plurality of keywords stored in the image storage unit from newly input text;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A combination of a plurality of keywords associated with the image extracted by the image extraction unit, and a combination weight of a plurality of keywords that is the same as a combination of the plurality of keywords stored in the combination weight storage unit, A weight calculation unit that calculates a total value of weights for each image extracted by the image extraction unit;
A predetermined number of images from an image having a large total weight of combinations of a plurality of keywords calculated by the weight calculation unit, or a total value of weights of a plurality of keyword combinations calculated by the weight calculation unit is predetermined. An image processing apparatus comprising: an image selection unit that selects an image larger than the obtained value from the images extracted by the image extraction unit. An image storage stage for storing the image in the image storage unit in association with a keyword indicating the content of the image;
A combination weight storage stage for storing the weight of the combination of the plurality of keywords in the combination weight storage unit in association with the combination of the plurality of keywords stored in the image storage unit;
An image keyword extraction step for extracting a plurality of keywords stored in the image storage unit from newly input text;
An image extraction step of extracting an image stored in the image storage unit in association with at least one keyword extracted in the image keyword extraction step;
A combination of a plurality of keywords associated with the image extracted in the image extraction step, and a combination weight of the plurality of keywords that is the same as the combination of the plurality of keywords stored in the combination weight storage unit, A weight calculation step of calculating a total weight value for each image extracted by the image extraction step;
A predetermined number of images from images having a large total weight of combinations of a plurality of keywords calculated by the weight calculation step, or a total value of weights of a plurality of keyword combinations calculated by the weight calculation step is predetermined. An image selection method comprising: an image selection step of selecting an image larger than the obtained value from the images extracted by the image extraction step. An image processing program for an image processing apparatus for extracting an image corresponding to a keyword included in an input sentence, the image processing apparatus comprising:
An image storage unit for storing an image in association with a keyword indicating the content of the image;
A combination weight storage unit that stores weights of combinations of the plurality of keywords in association with combinations of the plurality of keywords stored in the image storage unit;
An image keyword extraction unit that extracts a plurality of keywords stored in the image storage unit from newly input text;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A combination of a plurality of keywords associated with the image extracted by the image extraction unit, and a combination weight of a plurality of keywords that is the same as a combination of the plurality of keywords stored in the combination weight storage unit, A weight calculation unit that calculates a total value of weights for each image extracted by the image extraction unit;
A predetermined number of images from an image having a large total weight of combinations of a plurality of keywords calculated by the weight calculation unit, or a total value of weights of a plurality of keyword combinations calculated by the weight calculation unit is predetermined. An image processing program for causing an image larger than the obtained value to function as an image selection unit that selects an image extracted by the image extraction unit. A sentence storage unit storing a plurality of sentences in advance;
A sentence keyword weight storage unit that stores the weight of the keyword in association with a keyword included in the sentence stored in the sentence storage unit;
An image storage unit for storing an image in association with a keyword indicating the content of the image;
An image keyword extraction unit that extracts a keyword stored in the image storage unit from a newly input sentence;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A keyword stored in the image storage unit in association with the image extracted by the image extraction unit for each at least one image extracted by the image extraction unit, the keyword extracted by the image keyword extraction unit; A weight calculation unit that calculates a total value of keyword weights stored in the sentence keyword weight storage unit in association with the same keyword;
A predetermined number of images from an image having a large total weight value calculated by the weight calculation unit or an image in which the total weight value calculated by the weight calculation unit is larger than a predetermined value An image processing apparatus comprising: an image selection unit that selects an image extracted from the image. A sentence keyword weight storage stage for storing the weight of the keyword in the sentence keyword weight storage unit in association with keywords included in a plurality of sentences stored in advance in the sentence storage unit;
An image storage stage for storing the image in the image storage unit in association with a keyword indicating the content of the image;
An image keyword extraction step for extracting a keyword stored in the image storage unit from a newly input sentence;
An image extraction step of extracting an image stored in the image storage unit in association with at least one keyword extracted in the image keyword extraction step;
For each at least one image extracted in the image extraction step, a keyword stored in the image storage unit in association with the image extracted in the image extraction step, the keyword extracted in the image keyword extraction step; A weight calculation stage for calculating a total value of keyword weights stored in the sentence keyword weight storage unit in association with the same keyword;
The image extracting step calculates a predetermined number of images from an image having a large total weight value calculated by the weight calculating step, or an image having a total weight value calculated by the weight calculating step larger than a predetermined value. An image processing method comprising: an image selection step of selecting from images extracted from the image. An image processing program for an image processing apparatus for extracting an image corresponding to a keyword included in an input sentence, the image processing apparatus comprising:
A sentence storage unit storing a plurality of sentences in advance;
A sentence keyword weight storage unit that stores the weight of the keyword in association with a keyword included in the sentence stored in the sentence storage unit;
An image storage unit for storing an image in association with a keyword indicating the content of the image;
An image keyword extraction unit that extracts a keyword stored in the image storage unit from a newly input sentence;
An image extraction unit for extracting an image stored in the image storage unit in association with at least one keyword extracted by the image keyword extraction unit;
A keyword stored in the image storage unit in association with the image extracted by the image extraction unit for each at least one image extracted by the image extraction unit, the keyword extracted by the image keyword extraction unit; A weight calculation unit that calculates a total value of keyword weights stored in the sentence keyword weight storage unit in association with the same keyword;
A predetermined number of images from an image having a large total weight value calculated by the weight calculation unit or an image in which the total weight value calculated by the weight calculation unit is larger than a predetermined value Is an image processing program that functions as an image selection unit that selects from images extracted by.

Images