# US 9,436,891 B2 · Discriminating synonymous expressions using images > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,436,891 B2 - Document ID: US9436891B2 - Status: Granted - Kind: B2 - Application number: 13/954,645 - Filed: 2013-07-30 - Published: 2015-02-05 - Granted: 2016-09-06 - Earliest claimed: 2013-07-30 - Assignees: Globalfoundries Inc. - Applicants: Globalfoundries Inc. - Inventors: Chiaki Oishi; Tetsuya Nasukawa; Shoko Suzuki - Web page: https://patentgrep.com/patent/US9436891B2 ## CPC - G06F 40/247: Thesauruses; Synonyms - G06F 18/22: Matching criteria, e.g. proximity measures - G06V 10/758: {Involving statistics of pixels or of feature values, e.g. histogram matching} - G06V 10/768: {using context analysis, e.g. recognition aided by known co-occurring patterns} - G06V 10/462: {Salient features, e.g. scale invariant feature transforms [SIFT]} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method for identifying synonymous expressions includes determining synonymous expression candidates for a target expression. A plurality of target images related to the target expression and a plurality of candidate images related to each of the synonymous expression candidates are identified. Features extracted from the plurality of target images are compared with features extracted from the plurality of candidate images using a processor to identify a synonymous expression of the target expression. ## Claims ### Claim 1 A method for identifying synonymous expressions, comprising: determining synonymous expression candidates for a target non-facial-based expression; identifying a plurality of target images related to the target non-facial-based expression and a plurality of candidate images related to each of the synonymous expression candidates; and comparing features extracted from the plurality of target images with features extracted from the plurality of candidate images using a processor to identify a synonymous expression of the target non-facial-based expression, wherein the synonymous expression includes at least one of a word, a phrase, and a sound. ### Claim 2 (depends on 1) The method as recited in claim 1, wherein comparing includes employing a metric to compare features extracted from the plurality of target images with features extracted from the plurality of candidate images. ### Claim 3 (depends on 2) The method as recited in claim 2, wherein the metric includes at least one of a similarity metric and a distance metric. ### Claim 4 (depends on 2) The method as recited in claim 2, wherein comparing further comprises identifying the synonymous expression of the target non-facial-based expression based on the metric. ### Claim 5 (depends on 4) The method as recited in claim 4, wherein comparing further comprises ranking the images based on the metric and selecting a synonymous expression candidate associated with a most frequent candidate image as the synonymous expression. ### Claim 6 (depends on 5) The method as recited in claim 5, wherein selecting includes selecting the synonymous expression candidate associated with the most frequent candidate image above a threshold as the synonymous expression. ### Claim 7 (depends on 1) The method as recited in claim 1, wherein the features extracted from the plurality of target images and the features extracted from the plurality of candidate images includes color information. ### Claim 8 (depends on 1) The method as recited in claim 1, further comprising removing outliers from the plurality of target images. ### Claim 9 A non-transitory computer readable storage medium comprising a computer readable program for identifying synonymous expressions, wherein the computer readable program when executed on a computer causes the computer to perform the steps of: determining synonymous expression candidates for a target non-facial-based expression; identifying a plurality of target images related to the target non-facial-based expression and a plurality of candidate images related to each of the synonymous expression candidates; and comparing features extracted from the plurality of target images with features extracted from the plurality of candidate images to identify a synonymous expression of the target non-facial-based expression, wherein the target non-facial-based expression and the synonymous expression include at least one of a word, a phrase, and a sound. ### Claim 10 (depends on 9) The non-transitory computer readable storage medium as recited in claim 9, wherein comparing includes employing a metric to compare features extracted from the plurality of target images with features extracted from the plurality of candidate images, the metric including at least one of a similarity metric and a distance metric. ### Claim 11 (depends on 9) The non-transitory computer readable storage medium as recited in claim 9, wherein the features extracted from the plurality of target images and the features extracted from the plurality of candidate images includes color information. ### Claim 12 A system for identifying synonymous expressions, comprising: a candidate identifier determining synonymous expression candidates for a target non-facial-based expression; an image selector identifying a plurality of target images related to the target non-facial-based expression and a plurality of candidate images related to each of the synonymous expression candidates; and a comparator comparing features extracted from the plurality of target images with features extracted from the plurality of candidate images using a processor to identify a synonymous expression of the target non-facial-based expression, wherein the synonymous expression includes at least one of a word, a phrase, and a sound. ### Claim 13 (depends on 12) The system as recited in claim 12, wherein the comparator employs a metric to compare features extracted from the plurality of target images with features extracted from the plurality of candidate images. ### Claim 14 (depends on 13) The system as recited in claim 13, wherein the metric includes at least one of a similarity metric and a distance metric. ### Claim 15 (depends on 13) The system as recited in claim 13, wherein the comparator identifies the synonymous expression of the target non-facial-based expression based on the metric. ### Claim 16 (depends on 15) The system as recited in claim 15, wherein the comparator ranks the images based on the metric and select a synonymous expression candidate associated with a most frequent candidate image as the synonymous expression. ### Claim 17 (depends on 16) The system as recited in claim 16, wherein the comparator selects the synonymous expression candidate associated with the most frequent candidate image above a threshold as the synonymous expression. ### Claim 18 (depends on 12) The system as recited in claim 12, wherein the features extracted from the plurality of target images and the features extracted from the plurality of candidate images includes color information. ### Claim 19 (depends on 12) The system as recited in claim 12, wherein the comparator removes outliers from the plurality of target images. ### Claim 20 (depends on 1) The method as recited in claim 1, wherein the synonymous expression includes the sound.