Image retrieving method and apparatus that calculates characteristic amounts of data correlated with and identifying an image

Claim 2

An image retrieval method, comprising the steps of: storing in a data base a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data correlated with, and identifying each of the images; inputting additional-information data and illustrative image data for identifying a desired image stored in said data base; selecting additional information of an image by comparing the input additional information with the additional information of each image stored in the data base; generating a plurality of characteristic-amounts of data from, and correlated with the input illustrative image data; calculating the difference between each of the generated plurality of characteristic-amounts of data and characteristic-amounts of data of image data having the additional information selected in the selecting step; and selecting image data having a calculated difference in the order of smallest difference as the desired image.

Claim 3

A method according to claim 1, further comprising the steps of: selecting image data having a difference smaller than a predetermined value; sorting the image data selected in the selecting step in the order of smallest difference; and displaying the sorted image data.

Claim 4

An image retrieving apparatus for retrieving a desired image from a data base, comprising: a storage means in said data base for storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data correlated with, and identifying each of the images; a first input means for inputting additional-information data for identifying a desired image stored in said data base; a second input means for inputting an illustrative image of the desired image; additional-information selection means for selecting additional information of an image by comparing the input additional information with additional information of each image stored in the data base; generation means for generating a plurality of characteristic-amounts of data from, and correlated with the input illustrative image data; difference calculation means for calculating the difference between each of the generated plurality of characteristic-amounts of data and characteristic-amounts of data of image data having the additional information selected by said additional-information selection means; and selection means for selecting image data having a calculated difference in the order of smallest difference as the desired image.

Claim 5

An image retrieving apparatus for retrieving a desired image from a data base, comprising: a storage means in said data base for storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data correlated with, and identifying each of the images; a first input means for inputting additional-information data for identifying a desired image stored in said data base; a second input means for inputting an illustrative image of the desired image; additional-information selection means for selecting additional information of an image by comparing the input additional information with additional information of each image stored in the data base; generation means for generating a plurality of characteristic-amounts of data from, and correlated with the input illustrative image data; difference calculation means for calculating the difference between each of the generated plurality of characteristic-amounts of data and characteristic-amounts of data of image data having the additional information selected by said additional-information selection means; and selection means for selecting image data having a calculated difference in the order of smallest difference as the desired image.

Claim 6

An apparatus according to claim 3, further comprising: image-data selection means for selecting image data having a difference smaller than a predetermined value; means for sorting the image data selected by said image-data selection means in the order of smallest difference; and display means for displaying the sorted image data.

Claim 7

An image-retrieving system for retrieving a desired image from a data base, comprising: a storage means in said data base for storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data correlated with, and identifying each of the images; a first input means for inputting additional-information data for identifying a desired image stored in said data base; additional-information selection means for selecting a set of candidates as the desired image by comparing the input additional-information with the stored additional-information for each of the stored plurality of images; a second input means for inputting an illustrative image of the desired image; generation means for generating characteristic-amounts of data from, and correlated with the input illustrative image; similarity calculation means for calculating the similarity between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the input illustrative image; and selection means for selecting as the desired image a one of the candidates whose stored characteristic-amounts of data is most similar to the generated characteristic-amounts of data of the input illustrative image.

Claim 8

An image-retrieving system for retrieving a desired image from a data base, comprising: a storage means in said data base for storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data correlated with, and identifying each of the images; a first input means for inputting additional-information data for identifying a desired image stored in said data base; additional-information selection means for selecting a set of candidates as the desired image by comparing the input additional-information with the stored additional-information for each of the stored plurality of images; a second input means for inputting an illustrative image of the desired image; generation means for generating characteristic-amounts of data from, and correlated with the input illustrative image; similarity calculation means for calculating the similarity between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the input illustrative image; and selection means for selecting as the desired image a one of the candidates whose stored characteristic-amounts of data is most similar to the generated characteristic-amounts of data of the input illustrative image.

Claim 9

The image retrieving system of claim 5, wherein said similarity calculation means calculates similarity based on the difference between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the illustrative image.

Claim 10

The image retrieving system of claim 5, wherein said selection means selects the desired image based on the smallest difference between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the illustrative image.

Claim 11

The image retrieving system of claim 7, wherein said selection means includes: means for sorting the candidates based in the order of smallest difference between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the illustrative image; and display means for displaying the sorted candidates so as user may select one of the candidates as the desired image.

Claim 12

An image-retrieving method for retrieving a desired image from a data base, comprising the steps of: storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data for identifying each of the images; inputting additional-information data correlated with, and identifying a desired image stored in said data base; selecting a set of candidates as the desired image by comparing the input additional-information with the stored additional-information for each of the stored plurality of images; inputting an illustrative image of the desired image; generating characteristic-amounts of data from, and correlated with the input illustrative image; calculating the similarity between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the input illustrative image; and selecting as the desired image a one of the candidates whose stored characteristic-amounts of data is most similar to the generated characteristic-amounts of data of the input illustrative image.

Claim 13

An image-retrieving method for retrieving a desired image from a data base, comprising the steps of: storing a plurality of images, additional-information for identifying each of the plurality of images and calculated characteristic-amounts of data for identifying each of the images; inputting additional-information data correlated with, and identifying a desired image stored in said data base; selecting a set of candidates as the desired image by comparing the input additional-information with the stored additional-information for each of the stored plurality of images; inputting an illustrative image of the desired image; generating characteristic-amounts of data from, and correlated with the input illustrative image; calculating the similarity between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the input illustrative image; and selecting as the desired image a one of the candidates whose stored characteristic-amounts of data is most similar to the generated characteristic-amounts of data of the input illustrative image.

Claim 14

The image retrieving method of claim 9, wherein calculating the similarity is based on the difference between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the illustrative image.

Claim 15

The image retrieving method of claim 10, wherein selecting the desired image is based on the smallest difference between the stored characteristic-amounts of data of each of the set of candidates and the generated characteristic-amounts of data of the illustrative image.

Claim 16

The image retrieving method of claim 11, wherein the step of selecting the desired image includes the steps of: sorting the candidates based in the order of smallest difference between the stored characteristic-amounts of data of the set of candidates and the generated characteristic-amounts of data of the illustrative image; and displaying the sorted candidates so a user may select one of the candidates as the desired image.