Impact of Generative Modeling for Fundus Image Augmentation With Improved and Degraded Quality in the Classification of Glaucoma.
Ricardo LeonardoJoão GonçalvesAndré V. CarreiroBeatriz SimõesTiago OliveiraFilipe SoaresPublished in: IEEE Access (2022)
Keyphrases
- image classification
- image data
- input image
- image content
- multiscale
- fundus images
- image features
- single image
- preprocessing stage
- image analysis
- segmentation method
- blood vessels
- generative model
- hyperspectral images
- image representation
- classification accuracy
- thematic mapper
- image regions
- high quality
- test images
- region of interest
- support vector machine
- discriminative learning
- pattern recognition
- feature extraction
- decision trees
- age related macular degeneration
- degraded images
- feature selection
- unsupervised learning
- text classification
- image set
- feature points
- high resolution
- textural features
- feature vectors
- training set
- feature space
- training data