Research On Rule-Based Spatial Position Description Of Image

Spatial position description is an important research direction of spatial information field, it is widely used in many research areas such as machine vision, text-to-scene conversions, image retrieval and geographic information systems etc. Spatial direction relation is an important part of spatial position, the construction, improvement and systematic of its model have important theoretical and practical value in related research areas.This thesis summarizes the background and the current situation of the spatial position description, and presents concept, nature and feature of spatial direction relationship, also analyzes formal model of spatial direction relationship; According to the nature and character of the spatial direction relationship, it presents the spatial direction relationship classification method and reference framework method and analyzes the spatial direction relationship models and its applications; In the observer reference framework, it designs two spatial direction relationship models– viewing angle model and visual field histogram model, and uses two different types of description of spatial direction relationship; Finally, the thesis combines spatial direction relationship model with the length relationship, topological relationship model to apply to automobile auxiliary driving, and uses natural language generation method based on fuzzy rule to generate the natural language description of spatial position, the experimental result shows that the described method is feasible and consistent with human cognition.The study and result indicates that the method of spatial position description using spatial position description in spatial direction relationship model of observer reference framework and nature language generation of fuzzy rule is feasible, with the advantages of acknowledgement, and can be applied in vehicle driver assistance and robot vision etc.