| CPC G06F 21/10 (2013.01) [G06F 16/783 (2019.01); G06T 3/40 (2013.01); G06V 10/44 (2022.01); G06V 10/761 (2022.01); H04N 21/8352 (2013.01)] | 9 Claims |
|
1. A method for still image-based media content monitoring comprising:
an extraction step of receiving, by a video registration unit, an original content which is a video content and extracting a plurality of still images from the original content according to a predetermined condition;
a registration step of generating, by the video registration unit, an original frame by converting the extracted still image to a preset standard, generating feature point information by extracting a feature point from the original frame, and then matching content identification information and an original frame of the original content corresponding to the feature point information and storing the matched information in a database;
a reception step of extracting, by a video identification unit, a still image from a search target video at a preset time interval when receiving the search target video and converting the extracted still image from the search target video into the preset standard to generate a plurality of search target frames;
a comparison step of performing, by the video identification unit, an operation process of extracting feature points from the search target frame and then comparing the extracted feature points from the search target frame with the feature point information stored in the database for each original content to calculate visual similarity to the search target frame for each of one or more original frames matched at a reference or more and generating comparison result information by collecting original frames having visual similarity equal to or greater than a preset first reference value for each of the plurality of search target frames; and
a detection step of calculating, by a detection unit, an identification number by summing a number of comparison result information in which an original frame corresponding to an original content exists among a plurality of comparison result information corresponding to the plurality of search target frames for each different original content and detecting an original content matching the search target video based on the identification number, and
wherein the reception step further comprises
generating, by the video identification unit, reference feature point information by extracting a still image when receiving the search target video and then extracting a feature point from the extracted still image from the search target video, identifying an original content matching feature point information most matching the reference feature point information while matching the reference feature point information at a predetermined second reference value or more by comparing the reference feature point information with feature point information stored in the database, identifying the search target video as a trailer when the identified original content is the trailer, and identifying the search target video as a non-trailer when the identified original content is not the trailer; and
extracting, by the video identification unit, still images per second in a predetermined time section based on the still image from the original content used when identifying the search target video as the trailer by targeting the search target video when the search target video is not the trailer.
|