Wavelet Hash
Wavelet hashing hashes wavelet transform coefficients as signal features, used for multimedia perceptual hashing and content recognition.
Purpose
Provides perceptual hashing robust to common processing (compression, scaling, filtering) for multimedia retrieval.
Mechanism
The discrete wavelet transform (DWT) is applied to the signal; coefficients of energy-concentrated low-frequency subbands are selected, quantized, and encoded as the hash. Different wavelet bases (Daubechies, Coiflet) suit different signals.
Use cases
- Audio perceptual hashing
- Video fingerprinting
- Multimedia retrieval