# US 2018/0025227 A1 · Methods and apparatus to measure brand exposure in media streams > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0025227 A1 - Document ID: US20180025227A1 - Status: Published application - Kind: A1 - Application number: 15/721,171 - Filed: 2017-09-29 - Published: 2018-01-25 - Earliest claimed: 2007-11-09 - Assignees: The Nielsen Company (US), LLC - Applicants: The Nielsen Company (US), LLC - Inventors: Kevin Keqiang Deng; Barry Greenberg - Web page: https://patentgrep.com/patent/US20180025227A1 ## CPC - G06V 20/40: in video content (extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44) - G06V 20/635: {Overlay text, e.g. embedded captions in a TV programme} - G06V 20/44: {Event detection} - G06V 2201/09: Recognition of logos - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Methods and apparatus to measure brand exposure in media streams are disclosed. Disclosed example apparatus include a scene detector to compare a signature of a detected scene of a media presentation with a library of signatures to identify a first reference scene, and a scene classifier to classify the detected scene as a scene of changed interest when a first region of interest in the detected scene does not include a first reference brand identifier included in a corresponding region of interest in the first reference scene. Disclosed example apparatus further include a graphical user interface to present the detected scene, prompt for selection of an area of the first region of interest in the detected scene, and compare the selected area to a library of reference brand identifiers to identify a second reference brand identifier included in the first region of interest in the detected scene. ## Claims ### Claim 1 An apparatus to detect brand exposures included in a media presentation, the apparatus comprising: a scene detector to compare a signature determined from a detected scene obtained from the media presentation with a library of reference signatures corresponding to reference media to identify a first reference scene representative of the detected scene; a scene classifier to classify the detected scene as a scene of changed interest in response to determining a first region of interest in the detected scene, which corresponds to a first region of interest specified in the first reference scene, does not include a first reference brand identifier specified as being included in the first region of interest in the first reference scene; and a graphical user interface (GUI) implemented by a processor to present the detected scene in response to the detected scene being classified as a scene of changed interest, the GUI to further prompt for selection of an area of the first region of interest in the detected scene and compare the selected area to a library of reference brand identifiers to identify a second reference brand identifier included in the first region of interest in the detected scene. ### Claim 2 (depends on 1) The apparatus of claim 1, wherein the scene classifier is to classify the detected scene as a scene of interest in response to determining the first region of interest in the detected scene includes the first reference brand identifier specified as being included in the first region of interest in the first reference scene. ### Claim 3 (depends on 1) The apparatus of claim 1, wherein the GUI is to determine the selected area of the first region of interest in the detected scene by: detecting a first point in the first region of interest in the detected scene that was marked in response to the prompt; beginning with the first point, comparing adjacent pixels in the first region of interest in the detected scene to identify a group of pixels in the first region of interest in the detected scene, the group of pixels including the first point; and determining a bounding shape that includes the group of pixels, the bounding shape representing the selected area of the first region of interest in the detected scene. ### Claim 4 (depends on 3) The apparatus of claim 3, wherein the bounding shape includes at least one of a polygon or an ellipse. ### Claim 5 (depends on 3) The apparatus of claim 3, wherein the GUI is to compare the adjacent pixels based on at least one of luminance or chrominance. ### Claim 6 (depends on 3) The apparatus of claim 3, wherein the bounding shape corresponds to a first region representing the selected area of the first region of interest in the detected scene, and the GUI is further to determine the selected area of the first region of interest in the detected scene by: detecting, after determining the bounding shape, a second point selected in the first region of interest; and adjusting the first region to determine a second region having a boundary including the second point, the second region to represent the selected area of the first region of interest in the detected scene. ### Claim 7 (depends on 1) The apparatus of claim 1, wherein the GUI is further to: display the media presentation; and pause the display of the media presentation to present the detected scene. ### Claim 8 An article of manufacture comprising computer readable instructions that, when executed, cause a process to at least: compare a signature determined from a detected scene obtained from a media presentation with a library of reference signatures corresponding to reference media to identify a first reference scene representative of the detected scene; classify the detected scene as a scene of changed interest in response to determining a first region of interest in the detected scene, which corresponds to a first region of interest specified in the first reference scene, does not include a first reference brand identifier specified as being included in the first region of interest in the first reference scene; and in response to the detected scene being classified as a scene of changed interest, (1) present the detected scene, (2) prompt for selection of an area of the first region of interest in the detected scene and (3) compare the selected area to a library of reference brand identifiers to identify a second reference brand identifier included in the first region of interest in the detected scene. ### Claim 9 (depends on 8) The article of manufacture of claim 8, wherein the instructions, when executed, further cause the processor to classify the detected scene as a scene of interest in response to determining the first region of interest in the detected scene includes the first reference brand identifier specified as being included in the first region of interest in the first reference scene. ### Claim 10 (depends on 8) The article of manufacture of claim 8, wherein the instructions, when executed, further cause the processor to determine the selected area of the first region of interest in the detected scene by: detecting a first point in the first region of interest in the detected scene that was marked in response to the prompt; beginning with the first point, comparing adjacent pixels in the first region of interest in the detected scene to identify a group of pixels in the first region of interest in the detected scene, the group of pixels including the first point; and determining a bounding shape that includes the group of pixels, the bounding shape representing the selected area of the first region of interest in the detected scene. ### Claim 11 (depends on 10) The article of manufacture of claim 10, wherein the bounding shape includes at least one of a polygon or an ellipse. ### Claim 12 (depends on 10) The article of manufacture of claim 10, wherein the instructions, when executed, cause the processor to compare the adjacent pixels based on at least one of luminance or chrominance. ### Claim 13 (depends on 10) The article of manufacture of claim 10, wherein the bounding shape corresponds to a first region representing the selected area of the first region of interest in the detected scene, and the instructions, when executed, further cause the processor to determine the selected area of the first region of interest in the detected scene by: detecting, after determining the bounding shape, a second point selected in the first region of interest; and adjusting the first region to determine a second region having a boundary including the second point, the second region to represent the selected area of the first region of interest in the detected scene. ### Claim 14 (depends on 8) The article of manufacture of claim 8, wherein the instructions, when executed, further cause the processor to: display the media presentation; and pause the display of the media presentation to present the detected scene. ### Claim 15 A system to detect brand exposures included in a media presentation, the system comprising: means for comparing a signature determined from a detected scene obtained from the media presentation with a library of reference signatures corresponding to reference media to identify a first reference scene representative of the detected scene; means for classifying the detected scene as a scene of changed interest in response to determining a first region of interest in the detected scene, which corresponds to a first region of interest specified in the first reference scene, does not include a first reference brand identifier specified as being included in the first region of interest in the first reference scene; and user interface means, responsive to classifying the detected scene as a scene of changed interest, for (1) presenting the detected scene, (2) prompting for selection of an area of the first region of interest in the detected scene and (3) comparing the selected area to a library of reference brand identifiers to identify a second reference brand identifier included in the first region of interest in the detected scene. ### Claim 16 (depends on 15) The system of claim 15, further including means for classifying the detected scene as a scene of interest in response to determining the first region of interest in the detected scene includes the first reference brand identifier specified as being included in the first region of interest in the first reference scene. ### Claim 17 (depends on 15) The system of claim 15, wherein the user interface means is further to determine the selected area of the first region of interest in the detected scene by: detecting a first point in the first region of interest in the detected scene that was marked in response to the prompt; beginning with the first point, comparing adjacent pixels in the first region of interest in the detected scene to identify a group of pixels in the first region of interest in the detected scene, the group of pixels including the first point; and determining a bounding shape that includes the group of pixels, the bounding shape representing the selected area of the first region of interest in the detected scene. ### Claim 18 (depends on 17) The system of claim 17, wherein the comparing of the adjacent pixels is based on at least one of luminance or chrominance. ### Claim 19 (depends on 17) The system of claim 17, wherein the bounding shape corresponds to a first region representing the selected area of the first region of interest in the detected scene, and the user interface means is further to determine the selected area of the first region of interest in the detected scene by: detecting, after determining the bounding shape, a second point selected in the first region of interest; and adjusting the first region to determine a second region having a boundary including the second point, the second region to represent the selected area of the first region of interest in the detected scene. ### Claim 20 (depends on 15) The system of claim 15, wherein the user interface means is further to: display the media presentation; and pause the display of the media presentation to present the detected scene.