Pattern augmentation for classifying handwritten japanese hiragana characters of 46 classes by using CNN pre-trained with object images.
Yoshihiro ShimaYuki OmoriYumi NakashimaMichio YasudaPublished in: ICMV (2018)
Keyphrases
- pre trained
- training data
- three dimensional
- lighting conditions
- image retrieval
- partial occlusion
- image set
- input image
- training examples
- pixel level
- target object
- human faces
- image features
- object recognition
- image matching
- small number
- character recognition
- chinese characters
- control signals
- data sets
- supervised learning
- image registration
- learning algorithm