粒子滤波算法 Resources

Showing items tagged with "粒子滤波算法"

This is a ready-to-run source program for video tracking implementation using particle filter algorithm, with pre-configured initial parameters. The core algorithm is computationally optimized with C++ code compiled into DLL via MEX files for MATLAB integration. Users can modify it for frame-difference based localization to avoid manual parameter adjustments when switching videos. The implementation demonstrates efficient particle filter application in visual tracking, featuring resampling mechanisms and likelihood estimation for target state prediction. Currently limited to AVI format input, users need to provide their own video files and update filename references in the video reading statements.

MATLAB 220 views Tagged
Download

Beginner-friendly MATLAB code for particle filtering with practical applications in target tracking, SLAM, and image matching. This initial release demonstrates basic particle tracking implementation, with follow-up versions planned for sequential particle processing.

MATLAB 247 views Tagged
Download

This program implements an innovative particle filter-based algorithm that integrates MCMC Bayesian Model Selection and Markov Chain Monte Carlo methodologies for target tracking applications. It effectively handles single-target tracking, multi-target tracking, and video-based target localization with superior nonlinear problem-solving capabilities compared to Kalman Filter, EKF, and UKF approaches. The implementation includes key components for particle weight updating, resampling mechanisms, and state estimation using Monte Carlo simulations. This valuable technical resource is now shared to foster collaborative development and mutual support within the research community.

MATLAB 229 views Tagged
Download

Particle filter algorithm, state noise, observation noise, particle swarm size, updating weights for each particle in the swarm, weight normalization, resampling. Implementation includes handling noise models, weight calculation based on likelihood functions, and systematic resampling techniques.

MATLAB 205 views Tagged
Download