首页> 外文会议>Signal Processing, Pattern Recognition, and Applications >ANALYTICAL COMPUTER CALCULATIONS IN IMAGE ANALYSIS: CORRECT READING OF RANDOM DISCRETE-POINT FIELDS
【24h】

ANALYTICAL COMPUTER CALCULATIONS IN IMAGE ANALYSIS: CORRECT READING OF RANDOM DISCRETE-POINT FIELDS

机译:图像分析中的分析计算机计算:随机离散点场的正确读取

获取原文

摘要

The present paper is devoted to questions arising during the registration of random two-dimensional discrete -point fields of a various physical nature. For determination of coordinates of point sources forming such image, the sizes of the integrating aperture of the photo-reading device (integrator) are chosen depending on required accuracy of measurement. In order to prevent errors, which can happen if some point objects (2 or greater) get to the driven aperture-"window", the multithreshold integrators are applied. To solve applied tasks, in which the similar schemes of registration are used, engineers and practical researchers need know reliability of the received data. Unfortunately, the exact analytical formulas describing probability of correct reading, are known only for the elementary case, when the integrator has one threshold level. The research of multithreshold reading leads up, as a rule, to difficult probability problems related to random division of an interval (a lot of them are not decided on today). In this paper the original nontraditional algorithms are offered, and its essence is, that the exact formulas for probability of correct multithreshold reading are obtained with the help of analytical computer transformations.
机译:本文致力于各种物理性质的随机二维离散点场的配准过程中出现的问题。为了确定形成这种图像的点源的坐标,根据所需的测量精度来选择光读取装置(积分器)的积分孔的尺寸。为了防止错误,如果某些点对象(2个或更多)到达驱动光圈“窗口”时可能发生,则应用多阈值积分器。为了解决应用相似注册方案的应用任务,工程师和实际研究人员需要了解所接收数据的可靠性。不幸的是,当积分器具有一个阈值水平时,描述正确读取概率的精确分析公式仅在基本情况下才知道。通常,对多阈值读取的研究导致了与区间随机划分有关的困难概率问题(今天尚未决定其中的许多问题)。本文提供了原始的非传统算法,其实质是借助分析计算机变换获得正确的多阈值读取概率的精确公式。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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