摘要

In this paper, we developed a technique of information declassification and reversion which is applicable for vector geo-data in GIS. Firstly, for information declassification, a set of interfering points were created using logistic chaotic system. Then, to control the extent of information interfering, some auxiliary points were inserted by user-defined step. Finally, after the process of coordinate reversion, the recognition and elimination of auxiliary points, the data can be restored to original state. To evaluate the effects of this method, we assess the consistency of shape patterns by graphic complexity, the consistency of topological spatial relationships based on 4-intersection model,and the consistency of spatial directional relationships by local directional relationships of adjacent points. This technique implements information declassification according to the precision that user demands with stochastic interference and better automation, and enables the consistency of shape patterns (98.93%), spatial topology relationships (99.71%) and spatial directional relationships (99.20%) before and after declassification. It can meet the declassification requirements of vector geo-data for the applications such as data sharing, secure transmission, and sealed storage.