Classification of Diabetic Retinopathy Images Using Multi-Class Multiple-Instance Learning Based on Color Correlogram Features.
Ragav VenkatesanParag S. ChandakkarBaoxin LiPublished in: CoRR (2017)
Keyphrases
- multi class
- multiple instance learning
- image features
- support vector machine
- image annotation
- feature vectors
- image classification
- class labels
- feature extraction
- classification accuracy
- feature selection
- feature set
- supervised learning
- color histogram
- input image
- image collections
- feature space
- image data
- image retrieval
- image database
- spatial information
- image analysis
- binary classification
- object recognition
- object detection
- image understanding
- pairwise
- segmentation algorithm
- image descriptors
- image regions
- low level
- support vector
- decision trees
- mean shift
- similarity measure
- image search
- machine learning
- data sets
- segmentation method
- knn
- image set
- computer vision