Human and DNN Classification Performance on Images With Quality Distortions: A Comparative Study.
Samuel F. DodgeLina J. KaramPublished in: ACM Trans. Appl. Percept. (2019)
Keyphrases
- image classification
- input image
- image data
- image database
- pattern recognition
- geometric distortions
- image features
- classification accuracy
- human observers
- ground truth
- object recognition
- image analysis
- imaging process
- human subjects
- image retrieval
- three dimensional
- image collections
- decision trees
- distorted images
- edge detection
- hyperspectral images
- pigmented skin lesions
- image processing
- text classification
- machine learning
- image distortion
- image matching
- image registration
- image set
- support vector machine
- feature vectors
- pixel values
- training process
- training data
- similarity measure
- handwritten digits
- feature selection
- skin lesion
- preprocessing stage
- feature points
- computer vision