Automatic adjustment of the pulse-coupled neural network hyperparameters based on differential evolution and cluster validity index for image segmentation.
Wilfrido Gómez-FloresJuanita Hernández-LópezPublished in: Appl. Soft Comput. (2020)
Keyphrases
- differential evolution
- hyperparameters
- pulse coupled neural network
- image segmentation
- evolutionary algorithm
- image processing
- optimization algorithm
- model selection
- cluster validity index
- cross validation
- particle swarm optimization
- bayesian inference
- parameter optimization
- fuzzy c means
- genetic algorithm
- random sampling
- bayesian framework
- markov random field
- maximum a posteriori
- prior information
- closed form
- maximum likelihood
- support vector
- incremental learning
- multiscale
- noise level
- multi sensor
- feature extraction
- multi objective
- parameter space
- computer vision
- incomplete data
- data sets
- active learning
- sample size
- em algorithm
- graph cuts
- object recognition
- missing values
- fitness function
- similarity measure
- energy function
- clustering algorithm