METHOD, APPARATUS AND PROGRAM FOR SAMPLING LEARNING TARGET FRAME IMAGE OF VIDEO FOR ARTIFICIAL INTELLIGENCE IMAGE LEARNING, AND IMAGE LEARNING METHOD THEREOF

摘要

Provided are a method, apparatus, and program for sampling a learning target frame image of a video for artificial intelligence image learning, and an image learning method thereof. The method for sampling a learning target frame image of a video for artificial intelligence image learning, which is executed by a computer, comprises: a raw video reception step of receiving a raw video for artificial intelligence image learning; a frame image extraction step of extracting a predetermined number of frame images per a predetermined time section from the received raw video; a learning target object detection step of detecting one or more predetermined types of one or more learning target objects on each frame image by using an object detection algorithm; a background removal step of removing a background except the learning target object on each frame image; an object movement amount measurement step of measuring a movement amount of each learning target object on an n^th (n is a natural number greater than two) frame image from which a background is removed; and a learning target frame image selection step of selecting the n^th frame image as a learning target frame image by comparing a measurement result of a movement amount of each of the one or more learning target objects detected on the n^th frame image with a predetermined reference.;COPYRIGHT KIPO 2020