An objective IQA framework based on noise clusters estimation for still images.
Naho ItoYasuhiro InazumiYuukou HoritaPublished in: GCCE (2017)
Keyphrases
- high contrast
- image database
- image noise
- three dimensional
- image features
- image data
- ground truth
- noise free
- test images
- distorted images
- image classification
- image structure
- object recognition
- input data
- input image
- image registration
- human perception
- gaussian noise
- image retrieval
- image regions
- imaging process
- pixel intensities
- image collections
- image analysis
- pixel values
- clustering algorithm
- estimated parameters
- similarity measure
- signal subspace