首页> 外文会议>International Conference on Smart Grid and Electrical Automation >Research on Moving Small Target Tracking Method in Low Contrast
【24h】

Research on Moving Small Target Tracking Method in Low Contrast

机译:低对比度运动小目标跟踪方法研究

获取原文

摘要

With the constant development of science and technology, the method of small target tracking is widely used in military and civilian fields. But in the tracking process, the small target image is affected by the small proportion of pixels. Especially in conditions of low contrast, the tracking process is easy to appear the error tracking or the loss of target. For improving the motion small target trace ability in conditions of low contrast, this paper proposed a fusion based on filtering and data association positioning framework. Through the construction of small target track plots under the condition of low contrast and observation probability model, with the introduction of multi feature concept and the design of small target flight trajectory similarity probability, the relevance of target tracking data was improved. Simulation results show that the fusion technical method based on filtering and data association positioning framework improves efficiency compared with the traditional small targets tracking algorithm, which saves the computing time. Through the improvement of the small target tracking in the low contrast, a large number of tracking particles need not be wasted in the useless area, which saves the cost and improves the tracking efficiency.
机译:随着科学技术的不断发展,小目标跟踪方法已广泛应用于军事和民用领域。但是在跟踪过程中,较小的目标图像会受到较小比例的像素的影响。特别是在对比度较低的情况下,跟踪过程容易出现误差跟踪或目标丢失。为了提高低对比度条件下的运动小目标跟踪能力,提出了一种基于滤波和数据关联定位框架的融合方法。通过在低对比度和观测概率模型的条件下构造小目标航迹图,通过引入多特征概念和小目标飞行轨迹相似概率设计,提高了目标跟踪数据的相关性。仿真结果表明,与传统的小目标跟踪算法相比,基于过滤和数据关联定位框架的融合技术方法提高了效率,节省了计算时间。通过在低对比度下改善小目标跟踪,不需要在无用区域浪费大量的跟踪粒子,节省了成本,提高了跟踪效率。

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