Integrated multilevel image fusion and match score fusion of visible and infrared face images for robust face recognition
Citations
895 citations
Cites background or methods from "Integrated multilevel image fusion ..."
...2005 [67, 70], 2006 [68] 2007 [64, 65], 2008 [66, 71]...
[...]
...A simple fusion strategy is to first extract Gabor and LBP features in the parallel way, and then fuse two kinds of features on feature level, matching score level, or decision level [65]-[66]....
[...]
...It was concluded in [64]-[66] that Gabor wavelet and LBP based features are mutually complementary, because LBP captures the local appearance detail; while Gabor-wavelets extract shape information over a broader range of scales....
[...]
849 citations
751 citations
Cites background from "Integrated multilevel image fusion ..."
...Other approaches have also been proposed recently which explore the fusion of face images captured under visible and infrared light spectrum to improve the performance of face recognition [255, 265-267]....
[...]
245 citations
187 citations
References
40,147 citations
"Integrated multilevel image fusion ..." refers methods in this paper
...SVM is widely used in classification problems because it is designed to circumvent the overfitting problem [20–22] and is generalized to optimally perform classification on new training data....
[...]
14,565 citations
Additional excerpts
...Product rule fusion [36] Sum rule fusion [36] SVM fusion [37] DST fusion [39] Proposed DSm fusion Notre Dame 95.07 97.12 97.46 97.58 98.82 Equinox 93.20 94.33 95.05 96.51 98.08 images are not fused....
[...]
...To compare the performance of the proposed DSm based match score fusion algorithm, we chose four existing fusion algorithms, Product rule [36], Sum rule [36], SVM fusion [37], and DST fusion [39]....
[...]
...Product rule and Sum rule are based on statistical rules, SVM fusion algorithm is a learning based algorithm, and DST fusion algorithm is based on evidence theory....
[...]
...Dempster Shafer theory (DST) [38] based fusion algorithm is one example of this paradigm in which uncertain and fuzzy information are efficiently fused....
[...]
...Existing statistical and learning based fusion algorithms including DST fusion algorithm do not account for the conflicting region....
[...]
6,384 citations
5,563 citations
"Integrated multilevel image fusion ..." refers background in this paper
...For example, changes in lighting affect the representation of visible spectrum images and can influence feature extraction....
[...]
4,773 citations
Additional excerpts
...Researchers in medical imaging have used mutual information based registration algorithms to effectively fuse images from different modalities such as CT and MRI [32,33]....
[...]