首页> 外国专利> MOVEMENT MEANS ESTIMATION MODEL GENERATION DEVICE, MOVEMENT MEANS ESTIMATION MODEL GENERATION METHOD, AND MOVEMENT MEANS ESTIMATION MODEL GENERATION PROGRAM

MOVEMENT MEANS ESTIMATION MODEL GENERATION DEVICE, MOVEMENT MEANS ESTIMATION MODEL GENERATION METHOD, AND MOVEMENT MEANS ESTIMATION MODEL GENERATION PROGRAM

机译:运动均值估计模型生成装置,运动均值估计模型生成方法以及运动均值估计模型生成程序

摘要

PROBLEM TO BE SOLVED: To enhance the estimation accuracy of movement means estimation by enabling the extraction of a feature amount which acquires sequence changes of a feature of an individual positioning point from a GPS track.SOLUTION: A sequence expression extraction unit 30 extracts sequence expression having information, such as a movement time to each positioning point on the basis of GPS track segments stored in DBs 10, 20, and stores the sequence expression in DBs 40, 50. A sequence feature amount extraction model generation unit 60 generates a sequence feature amount extraction model on the basis of stored data of the DBs 40, 50, and stores the sequence feature amount extraction model in a DB 70. A sequence feature amount extraction unit 80 extracts a sequence feature amount by using the sequence feature amount extraction model of the DB 70, and stores the sequence feature amount in a DB 90. A feature amount connection unit 120 stores a connection feature amount obtained by connecting the conventional basic feature amount stored in a DB 110 and the sequence feature amount stored in the DB 90 in a DB 130. An estimation model generation unit 140 generates an estimation model on the basis of the stored data of the DB 130, and stores the estimation model in a DB 150.SELECTED DRAWING: Figure 1
机译:解决的问题:为了提高运动的估计精度,通过使得能够提取从GPS轨迹获取单个定位点的特征的序列变化的特征量来进行估计。解决方案:序列表达式提取单元30提取序列表达式具有诸如基于存储在DB 10、20中的GPS轨迹段的到每个定位点的移动时间之类的信息,并且将序列表达式存储在DB 40、50中。序列特征量提取模型生成单元60生成序列特征。量提取模型基于DB 40、50的存储数据,并且将序列特征量提取模型存储在DB 70中。序列特征量提取单元80通过使用DB的序列特征量提取模型来提取序列特征量。 DB 70,并将序列特征量存储在DB 90中。特征量连接单元120存储通过连接存储在DB 110中的常规基本特征量和存储在DB 130中的DB 90中的序列特征量。估计模型生成单元140基于DB 130的存储数据生成估计模型,并存储DB 150中的估算模型。选择的图纸:图1

著录项

  • 公开/公告号JP2016099110A

    专利类型

  • 公开/公告日2016-05-30

    原文格式PDF

  • 申请/专利权人 NIPPON TELEGR & TELEPH CORP NTT;

    申请/专利号JP20140233204

  • 发明设计人 ENDO YUKI;TODA HIROYUKI;KOIKE YOSHIMASA;

    申请日2014-11-18

  • 分类号G01C21/26;G06N99;G01S19/42;G08G1/005;G06Q50/10;

  • 国家 JP

  • 入库时间 2022-08-21 14:45:46

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号