Proposal of PolSAR Land Classification Using Full-Learning Quaternion Convolutional Neural Networks.
Yuya MatsumotoRyo NatsuakiAkira HirosePublished in: IGARSS (2022)
Keyphrases
- convolutional neural networks
- learning algorithm
- supervised learning
- classification accuracy
- learning phase
- incremental learning
- pattern recognition
- machine learning
- discriminative learning
- text classification
- knowledge acquisition
- unsupervised learning
- learning systems
- multi category
- decision trees
- statistical learning
- pattern classification
- learning process
- decision rules
- classification rules
- frequency domain
- data sets
- feature extraction
- reinforcement learning
- fourier transform
- learning problems
- support vector machine svm
- online learning
- feature set
- preprocessing
- semi supervised
- training set
- active learning