基于线型识别与逐层构面的遥感解译数据循环赋值算法  

作　　者：白雪芹 李远华[1] 王瑞 叶梦莹 冶亚琼 甘海伶 

机构地区：[1]吉林大学地球探测科学与技术学院,吉林长春130026

出　　处：《科技创新与应用》2022年第6期63-67,72,共6页Technology Innovation and Application

基　　金：吉林大学大学生创新创业训练计划国家级项目(202010183267)。

摘　　要：为了提高遥感解译矢量成果要素的快速类型判断与属性赋值水平,文章采用C#+ArcEngine与OLEDB数据库融合的二次开发模式,首先调用ArcEngine函数读取预处理遥感解译矢量数据及相关信息,通过数据库查询将数据的“线型、颜色、字段值”等图形参数或属性信息与优先权模型进行比对,得到构建面图形顺序和属性值对应关系。然后利用循环算法调用属性赋值函数完成面要素的生成与属性自动赋值。最后,通过空间拓扑和依赖关系标定异常图斑并修正,最终实现解译数据的自动赋值。研究结果表明,将地物间以拓扑和依赖关系记录在线属性或其图形参数中,再利用数据库动态匹配查询,可以建立线面图元间的逻辑关系,为自动赋值奠定空间基础;根据解译对象的空间关系确定其逻辑排序,再采用“循环式、逐层式”图形构建与属性赋值方法,可以准确而又快速地完成数据生成与属性赋值;通过解译对象间的包含、相邻、逻辑接触关系,可以大致判断任意图斑分类的合理性,进而为标定异常图斑奠定基础。In order to improve the fast type judgment and attribute assignment level of vector result elements in remote sensing interpretation,this paper adopts the secondary development mode of the fusion of C#+ArcEngine and OLEDB database.Firstly,the ArcEngine function is called to read the preprocessed remote sensing interpretation vector data and related information,and the graphic parameters or attribute information such as"linetype,color,field value"of the data are compared with the priority model through database query.The corresponding relationship between the order of construction surface graphics and attribute values is obtained.Then the circular algorithm is used to call the attribute assignment function to complete the generation of surface features and automatic attribute assignment.Finally,the abnormal spots are calibrated and corrected by spatial topology and dependency,and the automatic assignment of interpretation data is realized.The results show that the online attributes or their graphic parameters are recorded by topology and dependency,and then the dynamic matching query of database is used to establish the logical relationship between line and surface elements,which lays a spatial foundation for automatic assignment.The logical order is determined according to the spatial relationship of the interpreted objects,and then the"circular and layer-by-layer"graphic construction and attribute assignment method are adopted,which can accurately and quickly complete the data generation and attribute assignment;by interpreting the inclusion,adjacency and logical contact relations between objects,the rationality of arbitrary speckle classification can be roughly judged,which lays a foundation for calibrating abnormal patches.

关 键 词：遥感解译数据 线型识别 逐层构面 循环赋值 

分 类 号：P231.5[天文地球—摄影测量与遥感]