特征参数 Resources

Showing items tagged with "特征参数"

This implementation provides a functional Gray-Level Co-occurrence Matrix (GLCM) with feature parameter calculation, developed for academic research purposes. The code generates GLCM and computes multiple texture features including: f1 (Angular Second Moment), f2 (Contrast), f3 (Correlation), f5 (Inverse Difference Moment), f6 (Sum Average), f7 (Sum Variance), f9 (Difference Average), and f10 (Difference Variance). Though not perfect, it has been successfully validated in practical applications.

MATLAB 253 views Tagged

Voice activity detection is a crucial preprocessing step for extracting feature parameters in speech compression applications, and this program provides a practical MATLAB implementation of this process with signal processing algorithm explanations for speech technology learners.

MATLAB 260 views Tagged

Perform preprocessing on a speech signal, including windowing, calculating zero-crossing rate, short-term energy, and extracting various speech signal feature parameters. This process involves time-domain analysis techniques commonly implemented using signal processing libraries.

MATLAB 274 views Tagged

Extraction of seven instantaneous information-based signal features implemented through MATLAB algorithms: maximum value of zero-centered normalized instantaneous amplitude power spectral density, standard deviation of zero-centered normalized instantaneous amplitude absolute value, standard deviation of absolute value of instantaneous phase nonlinear components for zero-centered non-weak signal segments, standard deviation of instantaneous phase nonlinear components for zero-centered non-weak signal segments, standard deviation of absolute value of instantaneous frequency for zero-centered normalized non-weak signal segments, maximum value of normalized instantaneous frequency power spectral density for a signal segment, and feature parameters derived from distinct XI-axis projection characteristics of QPSK and 16QAM signals.

MATLAB 250 views Tagged