# US 12,223,709 B2 · Methods for more effectively moderating one or more images and devices thereof > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,709 B2 - Document ID: US12223709B2 - Status: Granted - Kind: B2 - Application number: 17/006,402 - Filed: 2020-08-28 - Published: 2020-12-17 - Granted: 2025-02-11 - Earliest claimed: 2017-10-09 - Assignees: Webfurther, LLC - Applicants: WebFurther, LLC - Inventors: Jonathan Freger; Joshua Buxbaum - Web page: https://patentgrep.com/patent/US12223709B2 ## CPC - G06V 20/00: Scenes; Scene-specific elements (control of digital cameras H04N23/60) - G06F 16/51: Indexing; Data structures therefor; Storage structures - G06F 18/22: Matching criteria, e.g. proximity measures - G06F 18/241: relating to the classification model, e.g. parametric or non-parametric approaches - G06V 10/70: using pattern recognition or machine learning (optical pattern recognition or electronic computations therefor G06V10/88) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Methods, non-transitory computer readable media, and image moderation management computing apparatuses that receive a request that includes an image to be moderated. A percentage match of the received image against one or more computer vision models for one or more different categories is identified. A determination is made when the percentage match is within a range between customizable lower and upper thresholds for one or more of the different categories. Image moderation analysis data on the received image from one of one or more moderator computing devices is obtained when the percentage match is within the range. One or more stored rules on the received image are executed based on the obtained image moderation analysis data. ## Claims ### Claim 1 A method for moderating one or more images, the method comprising: identifying, by a computing apparatus, based on an automated image analysis a percentage match of an image against one or more computer vision models for one or more categories to moderate with respect to accessibility to view the image and which is based on a deep learning technique; determining, by the computing apparatus, when the identified percentage match is within a range between customizable lower and upper thresholds for the one or more of the categories to moderate with respect to accessibility to view the image; obtaining, by the computing apparatus, image moderation analysis data on adjusting the identified percentage match of the image in at least one of the categories to moderate with respect to accessibility to view the image from one of one or more moderator computing devices when the identified percentage match is determined to be within the range; adjusting, by the computing apparatus, the identified percentage match to an adjusted percentage match above the upper threshold or below the lower threshold based on the image moderation analysis data; and executing, by the computing apparatus, one of one or more stored rules from a database on the image based on whether the identified percentage match or the adjusted percentage match for the corresponding one of the one or more categories to moderate with respect to accessibility to view the image is above the upper threshold or below the lower threshold. ### Claim 2 (depends on 1) The method of claim 1 further comprising: updating, by the computing apparatus, the one or more computer vision models based on the obtained image moderation analysis data. ### Claim 3 (depends on 2) The method of claim 2, wherein the updating update is executed in a real time mode. ### Claim 4 (depends on 2) The method of claim 2, wherein the updating is executed in a batch mode. ### Claim 5 (depends on 1) The method of claim 1, further comprising: executing, by the computing apparatus, a different one of one or more stored rules on the image when the percentage match is above the upper threshold. ### Claim 6 (depends on 1) The method of claim 1, further comprising: executing, by the computing apparatus, another different one of one or more stored rules on the image when the percentage match is below the lower threshold. ### Claim 7 An image moderation management computing apparatus, comprising memory comprising programmed instructions stored thereon and one or more processors configured to be capable of executing the stored programmed instructions to: identify based on an automated image analysis a percentage match of an image against one or more computer vision models for one or more categories to moderate with respect to accessibility to view the image and which is based on a deep learning technique; determine when the identified percentage match is within a range between customizable lower and upper thresholds for the one or more of the categories to moderate with respect to accessibility to view the image; obtain image moderation analysis data on adjusting the identified percentage match of the image in at least one of the categories to moderate with respect to accessibility to view the image from one of one or more moderator computing devices when the identified percentage match is determined to be within the range; adjust the identified percentage match to an adjusted percentage match above the upper threshold or below the lower threshold based on the image moderation analysis data; and execute one of one or more stored rules from a database on the image based on whether the identified percentage match or the adjusted percentage match for the corresponding one of the one or more categories to moderate with respect to accessibility to view the image is above the upper threshold or below the lower threshold. ### Claim 8 (depends on 7) The apparatus of claim 7, wherein the one or more processors are further configured to be capable of executing the stored programmed instructions to: update the one or more computer vision models based on the obtained image moderation analysis data. ### Claim 9 (depends on 8) The apparatus of claim 8, wherein the update is executed in a real time mode. ### Claim 10 (depends on 8) The apparatus of claim 8, wherein the update is executed in a batch mode. ### Claim 11 (depends on 7) The apparatus of claim 7, wherein the one or more processors are further configured to be capable of executing the stored programmed instructions to: execute a different one of one or more stored rules on the image when the percentage match is above the upper threshold. ### Claim 12 (depends on 7) The apparatus of claim 7, wherein the one or more processors are further configured to be capable of executing the stored programmed instructions to: execute, by the computing apparatus, another different one of one or more stored rules on the image when the percentage match is below the lower threshold.