Research On Data Storage Structures And Query Methods Of 3D Indoor Map

Currently,electronic map has been an inseparable part of our daily lives.With the development of the technology of computer and geographic,plenty of studies have concentrated on the related technologies of the outdoor electronic map,but the researches of indoor map still stay at its early stages.An important condition for the rapid development of outdoor map is the efficient management and query of map data.Considering the large amount of map data,the method of segmenting is widely used for processing map data in order to reduce the storage units.Furthermore,different from the outdoor environment,in a relatively small indoor space,there are many objects with complicated relationship,resulting in an uneven distribution of indoor data and some tightly-coupled data.Therefore,both the sheet division model applied in two-dimensional scenes and the tree-based segmentation structure applied in three-dimensional scenes cannot be directly applied.The above two methods can't maintain the three-dimensional feature of map data,and may lead to inefficient data storage and query.To address these challenges,a three-dimensional indoor map data storage model is proposed based on octree structure,and then an efficient map query algorithm is given base on neighbor finding algorithm on the basis of the model.At first,due to the low efficiency of data storage in 3D indoor map,a data storage model based on octree scene segmentation structure is proposed.It can improve storage efficiency while ensuring the three-dimensional characteristics of map data.The model proposed adopts an octree structure that commonly used in the field of three-dimensional images,and the model adds addressing code to the nodes of the tree according to the regularity of the octree,marking each node uniquely.Meanwhile,in terms of indoor object data expression,the box method is used for the approximate expression.Based on that,owing to the characteristics of the data in the three-dimensional indoor map scene,the data storage model is improved based on traditional octree structure to equip itself with backtracking ability.So that it is able to solve the problem that the dividing line is within the range of object data,improving the whole storage efficiency.Secondly,aiming at the problem of excessive cost and excessive time for 3D indoor map data query,we propose a data query algorithm based on octree storage structure.Considering the characteristics of map data query and application,closed constraint condition is added tothe neighboring node query algorithm based on octree storage structure with the concept of“query a little,spread around”.The method mainly relies on the fact that many partitions exist in the indoor environment and each room is independent.Finally,experimental scene has been established by combining ArcGIS and SketchUP.The data storage model proposed in this paper is verified,and the time cost and expenses of octree structure used in the process of storage is compared with that of other segmentation structures.Meanwhile,the efficiency of the proposed query method is analyzed through the metrics of query cost and comprehensive query time.