A Novel Approach to Image Assessment by Seeking Unification of Subjective and Objective Criteria Based on Supervised Learning.
Pipei HuangShiyin QinDonghuan LuPublished in: MIRAGE (2011)
Keyphrases
- supervised learning
- image data
- multiscale
- image features
- input image
- template matching
- single image
- image classification
- image analysis
- image content
- edge detection
- subjective quality
- image pixels
- segmentation method
- image representation
- test images
- image segmentation
- hough transform
- higher order
- machine learning
- image set
- learning algorithm
- image matching
- human observers
- subjective evaluation
- high resolution
- object recognition
- similarity measure
- feature points
- image retrieval
- low level
- image collections
- evaluation criteria
- data sets
- spatial information
- segmentation algorithm
- training set
- medical images