矢量轨迹有损压缩余弦垂距判别法  

CVDD algorithm for vector trajectory lossy compression

作　　者：李升宏 耿生玲[1,2] 田立勤 李路加[1] 陈娜 林连海 LI Shenghong;GENG Shengling;TIAN Liqin;LI Lujia;CHEN Na;LIN Lianhai(School of Computer Science,Qinghai Normal University,Xining 810008,China;Institute of Plateau Science and Sustainable Development,Xining 810008,China;School of Computer Science,North China Institute of Science and Technology,Langfang 065201,China)

机构地区：[1]青海师范大学计算机学院,青海西宁810008 [2]高原科学与可持续发展研究院,青海西宁810008 [3]华北科技学院计算机学院,河北廊坊065201

出　　处：《西安邮电大学学报》2021年第6期72-81,共10页Journal of Xi’an University of Posts and Telecommunications

基　　金：国家自然科学基金项目(61862055);青海省重点研发与转化基金项目(2020-SF-140)。

摘　　要：在矢量轨迹处理中,为了获得有效的矢量轨迹,提出余弦垂距判别(Cosine Vertical Distance Discrimination,CVDD)算法。该算法按序将轨迹中连续的三元组构造球面三角形,通过计算中间元素与前后元素的距离值识别三元组为密集点集或稀疏点集,给出压缩判据。对不同的轨迹路径,通过计算中间元素的垂距,再用余弦值判断三元组的轨迹识别为直道或弯道,从而给出判据和中间元素压缩条件。将该算法与几种传统的算法进行多角度对比,结果分析可得,CVDD算法的压缩效果优于传统方法,并在曲线的形变和失真程度上得到了优化,该算法的复杂度低、通用性强,可扩展到实时压缩和多维空间压缩中。In vector trajectory processing,cosine vertical distance discrimination was proposed to obtain effective vector trajectory.The algorithm constructs spherical triangles from continuous triples in sequence,and identifies triples as dense or sparse points sets by calculating the distance between the intermediate elements and the elements before and after,and gives compression criteria.For different path paths,the vertical distance of the intermediate elements is calculated,and then the path of the triplet is identified as straight or curved by the cosine value,so as to give the criterion and the compression condition of the intermediate elements.The results show that the CVDD algorithm has better compression effect than the traditional method,and is optimized in the deformation and distortion degree of curves.The algorithm has low complexity and strong versatility,and can be extended to real-time compression and multidimensional space compression.

关 键 词：矢量数据压缩 有损压缩 余弦垂距判别 三元组 道格拉斯-普克算法 

分 类 号：TP310[自动化与计算机技术—计算机软件与理论]