Parameter training method for convolutional neural networks based on improved Hausdorff-like derivative.
Kai JiaZhe GaoShasha XiaoPublished in: Expert Syst. Appl. (2024)
Keyphrases
- computational cost
- synthetic data
- training phase
- dynamic programming
- objective function
- detection method
- convolutional neural networks
- pairwise
- cost function
- experimental evaluation
- high accuracy
- linear model
- training samples
- gaussian function
- improved algorithm
- parameter space
- classification method
- clustering method
- segmentation algorithm
- edge detection
- multiresolution
- preprocessing
- computational complexity