A Jungle Computing approach to common image source identification in large collections of images.
Ben van WerkhovenPieter HijmaCeriel J. H. JacobsJason MaassenZeno J. M. H. GeradtsHenri E. BalPublished in: Digit. Investig. (2018)
Keyphrases
- input image
- image data
- image collections
- image features
- test images
- image retrieval
- image classification
- original images
- high contrast
- image pixels
- region of interest
- pixel values
- image dataset
- annotated images
- image regions
- image analysis
- image set
- image content
- image processing algorithms
- segmented images
- image noise
- reference images
- image matching
- intensity values
- gray level images
- segmentation method
- pixel level
- image database
- extracted from images
- normalized correlation
- imaging process
- edge detection
- sample images
- image structure
- grey level
- pixel intensities
- lighting conditions
- feature points
- natural images
- relevant images
- contrast enhancement
- ccd camera
- blind deconvolution
- aerial images
- visual features
- segmentation algorithm
- synthesized images
- gray value
- image statistics
- web images
- pixel wise
- color histogram
- image representation
- keypoints
- vanishing points
- illumination conditions
- digital imaging
- image annotation
- geometric distortions
- labeled images
- relevance feedback
- digital photos
- satellite images
- bounding box
- image registration
- gray level
- multiscale
- image quality
- gaussian noise
- photo collections
- single image
- scale space
- degraded images
- high resolution
- text queries
- object recognition
- image blocks
- image segmentation