噪声 Resources

Showing items tagged with "噪声"

Wavelet Transform Threshold Denoising Method, first proposed by Professors Johnstone and Donoho in 1992, is a nonlinear denoising technique. It achieves near-optimal performance in terms of minimum mean square error while featuring the simplest implementation and minimal computational complexity. The core principle: orthogonal wavelet decomposition provides time-frequency localization, where signal components exhibit larger wavelet coefficients while noise distributes uniformly across high-frequency bands. Implementation involves threshold selection, coefficient shrinkage, and signal reconstruction - typically implemented using soft/hard thresholding functions in wavelet toolkits.

MATLAB 197 views Tagged

Greatest Of Constant False Alarm Rate Detection - GO-CFAR Detector: Includes simulation of GO-CFAR detection for target-free signals with noise only, and GO algorithm simulation for target-containing signals in Rayleigh clutter background. Some programs primarily focus on plotting, displaying signal waveforms during GO-CFAR detection and threshold comparison diagrams; while others are designed to calculate detection probability and false alarm probability during the GO-CFAR detection process.

MATLAB 214 views Tagged

Application Context: Stochastic Resonance (SR) utilizes noise to enhance signal detection. In bistable systems, parameters a and b in the Langevin equation critically impact system performance and require careful selection. The package includes two GA (Genetic Algorithm) implementation examples - one simplified and one advanced - demonstrating parameter optimization. Technical Innovation: Unlike conventional noise suppression methods, SR leverages environmental noise for signal amplification. The genetic algorithm systematically optimizes system parameters through fitness-based selection, crossover, and mutation operations.

MATLAB 241 views Tagged

This MATLAB program implements Wiener filtering for noise reduction, delivering superior performance compared to spectral subtraction by eliminating sharp noise artifacts without introducing musical noise.

MATLAB 182 views Tagged

Like traditional median filters, adaptive median filters employ a rectangular window region Sxy for processing. The key difference lies in the adaptive filter's ability to dynamically adjust (i.e., increase) the filter window size based on predefined conditions during operation. When the center pixel of the window is identified as noise, it gets replaced by the median value; otherwise, the original pixel value remains unchanged. The filter output subsequently replaces the value at coordinates (x,y), corresponding to the current window center position.

MATLAB 219 views Tagged