一种基于XGML的RVIS表示方法  

An VRIS Representation Method Based on XGML

作　　者：邱静[1] 袁家政[1] 

机构地区：[1]北京联合大学北京市信息服务工程重点实验室,北京100101

出　　处：《北京联合大学学报》2016年第4期33-40,共8页Journal of Beijing Union University

基　　金：国家自然科学基金项目(61372148;61271369);国家科技支撑课题(2015BAH55F03);北京联合大学"人才强校项目"(BPHR2014A04;BPHR2014E02)

摘　　要：提出了一种基于XGML的RVIS表示方法。首先,借鉴XML和SVG语法,设计了一套适合RVIS表示的XGML模型;其次,提取图像特征,转换为XGML模型中的图元数据描述集合;最后,对图像进行区域分裂合并操作,对不同区域进行处理,将得到的区域转换为XGML模型中的图元素描述集合。RVIS方法通过对不同的光栅化矢量图像进行试验,结果表明:此方法能够很好地将光栅化矢量图像表示为半结构化的XGML文档,使其存储容量减小且便于检索。We have proposed a method of VRIS representation based on XGML. We have designed an imitation XGML model based on SVG and XML grammar to assist our VRIS representation. In the processing of our method, image features are converted into a meta-data descriptor. Then, using a split-and-merge algorithm, transforming image into an element descriptors by utilizing XGML model. The model was applied to an experimental data set which includes various vector images. Experiment results show that our VRIS representation method can efficiently represent raster images which originally were vector image to a semi-structured XGML document, and it successfully reduces storage capacity and increases the speed of image retrieval.

关 键 词：XGML 半结构化 XML SVG 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]