The Visual Nearest Neighbor Query On Moving Object In Barriers Space

Spatial databases is a novel database system for storing and manipulating spatialdata, and support the corresponding retrieve function. There are many applicationsbased on it, such as geographic information systems, location-aware system, trafficdetection systems and environmental information systems. Space object query is themajor technology of spatial database. Although the space object query technology hasbeen widely used in real life, but still need to further improve query methods tocontinuously meet people’s new query requirements.This paper analyzes the nearest neighbor query technique on the R-tree, thenearest neighbor query is a very important operation of spatial data query, it has foundwide application since it was proposed. for example, the driver need to find thenearest gas station on the highway. Continuous nearest neighbor query is morecomplex queries, but in spatial data query technology is equally important, forexample, the driver of the car on the highway, he wanted to know the location of thenearest gas station on a section of road, the search result is a collection who includedifferent section and corresponding nearest gas station. But all these application didnot consider the possible presence of obstructions environment, the impact ofobstacles on our target queries in real life. We consider nearest neighbor querying in aenvironment exist continuous barrier, such as first-person shooter game, thecharacters need to shoot the target closest to him when moving, if the target isblocked by obstacle, then the role cannot shoot the target. So we need to calculate themoving track of figure and get the nearest visual target away from the role.In this paper, we do a research of visual nearest neighbor query methods. Firstly,the formal definition has been proposed, and reveal their unique trait. Then aneffective pruning strategies and rules to deal with problems encountered during thequery is proposed. For example, determine the intermediate nodes that has no effecton the query results, and make sure that the query won’t visit it in the process. Wealso propose an effective solution for quick check obstacles whether affect thevisibility of the query line. Finally, a variant of continuous visual nearest neighborquery, a nearest neighbor query method under limits of distance threshold. And usingthe real and synthetic data to verify and compare query results in the experimental part.