Method and apparatus for extracting spatio-temporal feature and detecting video copy based on the same in broadcasting communication system

A method for extracting spatio-temporal feature for video copy detection in a broadcasting communication system, comprising:

• dividing a video frame into a plurality of blocks and calculating an average luminance value of the blocks each having a luminance value, wherein video contents includes a plurality of video frames;
• applying a spatial weight to the average luminance value, to thereby generate a weighted average luminance value of the video frame;
• calculating differential luminance difference of the weighted average luminance value of the video frame;
• selecting the video frame having abrupt luminance change when a local threshold is applied to the differential luminance difference of the video frame, as a keyframe of the video contents;
• and
• calculating spatio-temporal feature of the video contents based on temporal and spatial characteristic of the keyframe.

A method for detecting video copy based on spatio-temporal feature in a broadcasting communication system, comprising:

• extracting keyframe feature of an original video;
• extracting keyframe feature of a query video;
• comparing the extracted keyframe feature of the original video and the extracted keyframe feature of the query video;
• and
• calculating similarity between the original video and the query video and determining whether the query video is a copy of the original video or not based on the calculated similarity.