A Method of Data Augmentation for Classifying Road Damage Considering Influence on Classification Accuracy.
Haruki TsuchiyaShinji FukuiYuji IwahoriYoshitsugu HayashiWitsarut AchariyaviriyaBoonserm KijsirikulPublished in: KES (2019)
Keyphrases
- classification accuracy
- synthetic data
- training data
- prior knowledge
- input data
- data sets
- significant improvement
- noisy data
- data collection
- preprocessing
- test data
- statistical methods
- segmentation method
- spatial data
- missing data
- prior information
- detection method
- information loss
- correlation analysis
- high accuracy
- probabilistic model
- pairwise
- data analysis
- model selection
- training samples
- support vector machine svm
- data mining techniques
- data structure
- feature selection
- learning algorithm
- machine learning
- feature subset
- data mining
- neural network