Similarity Retrieval on Video Database Based upon Module Operation

碩士 === 國立交通大學 === 資訊管理研究所 === 83 === In this thesis, we propose a method of retrieving videos from thevideo database based upon the temporal relationship among the videos. We transform each video(query) into a set of ordered triples, (Oi,Oj,Rij)s, where O...

Full description

Bibliographic Details
Main Authors: Huang, Lee-Wen, 黃立文
Other Authors: Lo, Chi-Chun
Format: Others
Language:zh-TW
Published: 1995
Online Access:http://ndltd.ncl.edu.tw/handle/95237395890701420282
Description
Summary:碩士 === 國立交通大學 === 資訊管理研究所 === 83 === In this thesis, we propose a method of retrieving videos from thevideo database based upon the temporal relationship among the videos. We transform each video(query) into a set of ordered triples, (Oi,Oj,Rij)s, where Oi and Oj are two symbol objects and Rij is the temporal relationship between Oi and Oj. Then, we construct a hashing table for all the triples corresponding to the videos in the video database. Every ordered triple is assigned a prime number. So, each video Vi can be transformed into a positive integer value Pi. A query can be transformed into a positive integer value Pq via the preconstructed hashing table. The answer to a query is a collection of videos such that their corresponding Pi''s can be divided by Pq, i.e., the remainder of Pi/Pq is equal to 0. A video database query system is built based upon the proposed method. From computation results, we notice that the query time increases linearly as the size of the video database grows.