Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features.
Ragav VenkatesanParag S. ChandakkarBaoxin LiHelen K. LiPublished in: EMBC (2012)
Keyphrases
- multi class
- multiple instance learning
- image features
- image annotation
- support vector machine
- image classification
- class labels
- supervised learning
- feature vectors
- classification accuracy
- feature set
- feature selection
- color histogram
- input image
- feature extraction
- feature space
- image database
- spatial information
- image data
- image analysis
- image retrieval
- image understanding
- pairwise
- object detection
- binary classification
- image set
- image descriptors
- image collections
- support vector
- image regions
- training samples
- visual features
- decision trees
- object recognition
- machine learning
- reinforcement learning
- segmentation algorithm
- image content
- semi supervised learning
- training set