Color Facial Image Representation with New Quaternion Gradients

This paper proposes a new color image representation and multiple feature fusion based method for improving color face recognition performance under different lighting conditions. First, a new image color image representation has been derived. Second, a quaternion gradient has been given to enhance...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Electronic Imaging Ročník 30; číslo 13; s. 383-1 - 383-7
Hlavní autoři: Grigoryan, Artyom, Agaian, Sos
Médium: Journal Article
Jazyk:angličtina
Vydáno: Society for Imaging Science and Technology 28.01.2018
Témata:
Color Facial Image Representation
Facial Image Processing
Local Binary Patterns
Quaternion Gradients
Visibility Images
ISSN:2470-1173, 2470-1173
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper proposes a new color image representation and multiple feature fusion based method for improving color face recognition performance under different lighting conditions. First, a new image color image representation has been derived. Second, a quaternion gradient has been given to enhance and extract the faces/object's edges, contours, and texture. Also, we propose a novel feature representation based on Quaternion Gradient-based LBP tool for color face recognition. Finally, we present a concept of combining the color facial recognition system, which is based on the local quaternion gradients based binary patterns LBP Image Representation, and a new color-to-gray new mapping. The presented concept can be used for surveillance, security systems, computer animation, face tagging, human-computer interface, biometric identification, behavioral analysis, content-based image and video indexing applications.
Bibliografie:2470-1173(20180128)2018:13L.3831;1-
ISSN:2470-1173
2470-1173
DOI:10.2352/ISSN.2470-1173.2018.13.IPAS-383