A novel two-level method for the computation of the LSP frequencies using a decimation-in-degree algorithm.
Chung-Hsien WuJau-Hung ChenPublished in: IEEE Trans. Speech Audio Process. (1997)
Keyphrases
- high accuracy
- detection algorithm
- dynamic programming
- computational cost
- preprocessing
- detection method
- improved algorithm
- cost function
- optimization algorithm
- objective function
- experimental evaluation
- significant improvement
- theoretical analysis
- synthetic and real images
- clustering method
- selection algorithm
- computational complexity
- recognition algorithm
- classification method
- support vector machine svm
- single pass
- high efficiency
- probabilistic model
- matching algorithm
- segmentation algorithm
- estimation algorithm
- similarity measure
- segmentation method
- parallel implementation
- convergence rate
- noisy data
- optimization method
- previously computed
- computational efficiency
- classification algorithm
- mathematical model
- computationally efficient
- region of interest
- prior information
- tree structure
- experimental study
- reconstruction method
- input data
- k means
- test images
- discrete fourier transform
- np hard
- energy function
- particle swarm optimization
- multiresolution
- block matching
- pairwise