Search criterion determination system, search system, and computer readable medium
A search criterion determination system includes: at least one processor configured to extract, in a case where plural images are inputted as a query, features from elements included in each image, determine at least one feature to be provided to a search engine as an input feature based on a relation between the extracted features, and transmit information based on the input feature to the search engine.
1. A search criterion determination system comprising:
at least one processor configured to:
extract, in a case where a plurality of images is inputted as a query, features from elements included in each image;
determine, based on a frequency of appearance of each feature in the extracted features, at least one first feature that (i) does not contradict another feature of the extracted features, or (ii) is consistent with the other feature of the extracted features, to be provided to a search engine as an input feature, wherein
the at least one first feature is determined by classifying the plurality of images into feature groups by type, and eliminating a feature having a frequency of appearance that is less than or equal to a threshold in each feature group; and
transmit information based on the input feature to the search engine.
2. The search criterion determination system according to claim 1 , wherein
the at least one processor determines, as another input feature, a second feature with a frequency of appearance or an appearance ratio higher than that of a different feature of the extracted features even though the second feature contradicts or is not consistent with the different feature of the extracted features.
3. The search criterion determination system according to claim 1 , wherein
the threshold has a different value depending on a category to which each of the features belongs.
4. The search criterion determination system according to claim 1 , wherein
the at least one processor presents, to a user, a second feature with a possibility of contradicting or being inconsistent with a different feature of the extracted features, and provides the second feature to the search engine in response to the second feature being selected by the user.
5. The search criterion determination system according to claim 1 , wherein
the at least first feature includes a plurality of features with a possibility of coexisting with each other, and the at least one processor transmits the plurality of features to the search engine.
6. The search criterion determination system according to claim 1 , wherein
the at least one processor associates the extracted features with a type of the images from which the features have been extracted.
7. The search criterion determination system according to claim 6 , wherein
the at least one processor presents, to a user, a screen for receiving a selection on whether to associate the extracted features with the type of the images from which the features have been extracted.
8. The search criterion determination system according to claim 1 , wherein
the at least one processor extracts an overall feature from each image, and provides the extracted overall feature to the search engine.
9. The search criterion determination system according to claim 8 , wherein
the at least one processor presents, to a user, a screen for receiving a selection on whether to provide the extracted overall feature of each image to the search engine.
10. A search system comprising:
at least one processor configured to:
extract, in a case where a plurality of images is inputted as a query, features from elements included in each image;
determine, based on a frequency of appearance of each feature in the extracted features, at least one feature that (i) does not contradict another feature of the extracted features, or (ii) is consistent with the other feature of the extracted features, to be provided to a search engine as an input feature, wherein
the at least one feature is determined by classifying the plurality of images into feature groups by type, and eliminating a feature having a frequency of appearance that is less than or equal to a threshold in each feature group;
transmit information based on the input feature to the search engine; and
search at least one image similar to the images inputted as the query, using the information based on the input feature.
11. A non-transitory computer readable medium storing a program causing a computer to execute a process comprising:
extracting, in a case where a plurality of images is inputted as a query, features from elements included in each image;
determining, based on a frequency of appearance of each feature in the extracted features, at least one feature that (i) does not contradict another feature of the extracted features, or (ii) is consistent with the other feature of the extracted features, to be provided to a search engine as an input feature, wherein
the at least one feature is determined by classifying the plurality of images into feature groups by type, and eliminating a feature having a frequency of appearance that is less than or equal to a threshold in each feature group; and
transmitting information based on the input feature to the search engine.