特征提取 Resources

Showing items tagged with "特征提取"

Image registration process using thin plate splines involves the following steps: First, feature extraction from both images to obtain feature points; then finding matching feature pairs through similarity measurement. Accurate feature extraction ensures successful feature matching. Therefore, finding feature extraction methods with good invariance and accuracy is crucial for matching precision. This implementation uses thin plate splines for image registration, which provides smooth deformation fields while minimizing bending energy.

MATLAB 218 views Tagged

This program designs and implements a basic speaker recognition system using MATLAB. Key functional modules include speech signal input management, feature extraction for speech discrimination, and speaker authentication. The system utilizes MFCC (Mel-Frequency Cepstral Coefficients) for feature extraction and GMM (Gaussian Mixture Models) for pattern matching, providing a comprehensive solution with additional modules for error handling and user interface.

MATLAB 280 views Tagged

Simulated Annealing Program: This code implements feature extraction in pattern recognition using the simulated annealing method. Users can easily enhance the algorithm's performance through modifications such as temperature schedule adjustments, memory mechanism integration, or combining with genetic algorithms.

MATLAB 202 views Tagged

This repository contains the official implementation from Dr. Lalonde's ECCV 2010 paper (Carnegie Mellon University), featuring comprehensive image processing workflows including image segmentation, feature extraction, and AdaBoost classification. The code employs sophisticated features such as color ratios, texture analysis, and skewness measurements, integrating road scene context for robust shadow detection. The implementation demonstrates high detection accuracy while requiring substantial computational time, making it valuable for shadow detection research and development.

MATLAB 214 views Tagged