A Database Selection Method for Cherry Blossom Movies Based on Changing Trend of Color Frequency

Abstract

In this paper, we propose a database selection method for cherry blossom movie databases constructed on the assumption that large amounts of movie data on different observation locations are constantly stored into the corresponding areas of the movie database. Our proposed method extracts a suitable candidate set of movie databases according to the user's image query based on the color information that summarizes the characteristics of each movie database in the distributed movie database environment. Our method focuses on the color frequency trend of cherry blossoms in a movie database to extract the movie databases that include flowering states in the image frames similar to the user's input cherry blossom image and to extract the period by analyzing the increasing and decreasing trend of the specified flower color within the specified period. In the experiment, we evaluate the feasibility of our proposed method in the distributed movie database environment, where each movie database consists of pseudo-movie data on cherry blossoms.