Abstract
This paper proposes an educational information system based on u-learning that supports two phase feature extract. A key frame selected by user in annotation-based retrieval takes query image of feature-based retrieval and by agent searches and displays the most similar key frame image after comparing query image with key frames in the DB through full image and center image, binary image histogram techniques which we proposed. Therefore, user can get more correct retrieval result by agent’s learning through repeated retrievals about video data for the environment education of infants. From experiment, the designed and implemented system shows increase of recall rate of about 1.5% and 2.0% as precision rate 94.2% for video data scene retrieval in performance assessment.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call