Stratified resampling strategy imported by residual information in SMC framework

doi:10.3778/j.issn.1002-8331.2010.21.008

Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (21): 30-33.DOI: 10.3778/j.issn.1002-8331.2010.21.008

• 博士论坛 • Previous Articles Next Articles

Stratified resampling strategy imported by residual information in SMC framework

WU Gang^1，2，TANG Zhen-min²，YANG Jing-yu²

1.Department of Vehicle Engineering，Nanjing Institute of Technology，Nanjing 211167，China
2.School of Computer Science & Technology，Nanjing University of Science and Technology，Nanjing 210094，China

Received:2010-04-14 Revised:2010-05-31 Online:2010-07-21 Published:2010-07-21
Contact: WU Gang

SMC框架中引入残差信息的分层重采样策略

吴刚^1，2，唐振民²，杨静宇²

1.南京工程学院车辆工程系，南京 211167
2.南京理工大学计算机科学与技术学院，南京 210094

通讯作者: 吴刚

Abstract

Abstract: Recently，sequential Monte Carlo theory has been applied abroad in different domains such as self-determined navigation，non-linear estimation and finance，and it attracts researchers more and more.A stratified resampling strategy is brought forward imported by residual information in this work.Accumulative distributing function is constructed by importing residuals of particles’ weights.Synchronously，sequential stochastic numbers are gradually produced by arranging on stochastic muster.The way improves the rationality and efficiency of sampling strategy.The method is confirmed by experiment based on emulational program.Compared with residual resampling strategy，polynomial resampling strategy and genetic resampling strategy，the systemic errors about resampling strategy are lowest in posterior mean error，mean square error and running time.The improved resampling strategy is embedded in object tracking algorithm.The correlative results show astringency and antinoise-capability about sampling strategy are excellent on data based on standard testing video.

Key words: sequential Monte Carlo, resampling, stratified, residual

摘要： 近年来序列蒙特卡罗理论及其应用在自动导航﹑非线性估计与金融等诸多领域受到了越来越广泛的关注。提出了一种引入残差信息的分层重采样策略，通过引入当前粒子集权值的残差来构建累积分布函数，同时针对随机区间逐级分层以产生有序的随机数集合，从而提高重采样的合理性与采样效率。首先从仿真实验的角度证明了它的有效性，对比残差重采样﹑多项式重采样与遗传重采样，提出的重采样策略在后验均值误差﹑均方差与运行时间方面均为最小；将提出的重采样策略嵌入到运动目标跟踪算法中，基于标准测试视频的跟踪结果同样佐证了该重采样策略的收敛性及良好的抗噪性能。

关键词: 序列蒙特卡罗, 重采样, 分层, 残差

CLC Number:

TP391.4

WU Gang^1，2，TANG Zhen-min²，YANG Jing-yu². Stratified resampling strategy imported by residual information in SMC framework[J]. Computer Engineering and Applications, 2010, 46(21): 30-33.

吴刚^1，2，唐振民²，杨静宇². SMC框架中引入残差信息的分层重采样策略[J]. 计算机工程与应用, 2010, 46(21): 30-33.

[1]	LI Xianguo, FENG Xinxin, LI Jianxiong. Sigle Image Super-Resolution Reconstruction Based on Multi-scale Residual Network [J]. Computer Engineering and Applications, 2021, 57(7): 215-221.
[2]	HU Yangtao, PEI Yang, LIN Chuan, LI Shicheng, YI Yugen. Atrous Residual U-Net for Retinal Vessel Segmentation [J]. Computer Engineering and Applications, 2021, 57(7): 185-191.
[3]	CHEN Renhe, LAI Zhenyi, QIAN Yurong. Improved Image Denoising Generative Adversarial Network Algorithm [J]. Computer Engineering and Applications, 2021, 57(5): 168-172.
[4]	JIN Wang, YI Guohong, HONG Hanyu, CHEN Siyuan. Real-Time Vehicle Detection Based on Convolutional Neural Network [J]. Computer Engineering and Applications, 2021, 57(5): 222-228.
[5]	XU Qihao, LI Bo. Method for Detecting Pulmonary Nodules Based on NRU Network [J]. Computer Engineering and Applications, 2021, 57(4): 83-90.
[6]	ZHOU Huaqiang, CAO Lin, DU Kangning. Sketch Face Synthesis Based on Multi-discriminator Cyclic Generative Adversarial Network [J]. Computer Engineering and Applications, 2021, 57(3): 231-238.
[7]	XIAO Ruixue, FENG Yingwei, QU Jianping. Steganalysis of Variable Size Image Based on Efficient Feature Fusion [J]. Computer Engineering and Applications, 2021, 57(24): 126-134.
[8]	HUANG Yinglai, AI Xin. Research on Classification of Corn Leaf Disease Image by Improved Residual Network [J]. Computer Engineering and Applications, 2021, 57(23): 178-184.
[9]	CHANG Hui, RAO Zhiqiang, ZHAO Yulin, LI Yichen. Research on Tunnel Crack Segmentation Algorithm Based on Improved U-Net Network [J]. Computer Engineering and Applications, 2021, 57(22): 215-222.
[10]	WANG Jia, ZHANG Nan, MENG Fanyun, WANG Jinhe. Improved Semantic Segmentation Algorithm Based on Pyramid Scene Parsing Network [J]. Computer Engineering and Applications, 2021, 57(19): 220-227.
[11]	LIU Xingchen, JIA Juncheng, ZHANG Li, HU Qinhan. Feature Concentration Network for Image Super-Resolution [J]. Computer Engineering and Applications, 2021, 57(16): 213-219.
[12]	XIA Mengqi, HAO Kun, ZHAO Lu. Monocular Image Depth Estimation Based on Fully Convolutional Encoder-Decoder Network [J]. Computer Engineering and Applications, 2021, 57(14): 231-236.
[13]	ZHANG Xiaoxiao, NIU Fu, MAO Jianping, AN Jubai, GUO Hao. Research on Identification of Sea Surface Oil Spill Zone Based on Multi-feature Fusion and Residual Network [J]. Computer Engineering and Applications, 2021, 57(14): 267-274.
[14]	ZUO Baochuan, ZHANG Qing. Feature Guidance Mechanism for Saliency Detection Network [J]. Computer Engineering and Applications, 2021, 57(14): 201-208.
[15]	CHEN Guihui, CHEN Wu, LI Zhongbing, YI Xin, LIU Huikang, HAN Chunyang. Image Super Resolution Reconstruction Based on Residual Convolution Attention Network [J]. Computer Engineering and Applications, 2021, 57(12): 193-200.

Stratified resampling strategy imported by residual information in SMC framework

SMC框架中引入残差信息的分层重采样策略

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics