Decision Fusion Based on Voting Scheme for IR and Visible Face Recognition

Shahbe, M. D.; Hati, S.

doi:10.1109/cgiv.2007.33

Cited by 3 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Decision-level fusion, where the core idea is to use different classifiers to vote on the final classification result, either by majority voting or by weighted voting. In [21], the method first reduces the dimensionality of the data in the dataset by Fisherfaces, then defines four distance metrics to measure the matching score between the test image and the training image, and finally assigns a weight to each matching score, and obtains the final matching score, i.e., the recognition result, by weighted voting.…”

Section: Feature Fusion Strategiesmentioning

confidence: 99%

Face Recognition Algorithm based on Optimal Fusion of LBP and HOG Features

2023

JISSR

View full text Add to dashboard Cite

Aiming at the problem that face recognition methods only rely on single feature recognition leading to low recognition accuracy, a face recognition algorithm based on the optimal fusion of LBP and HOG features is proposed. The method first extracts the face features through LBP and HOG operators and subsequently fuses the two features using linear weighting. Unlike traditional feature fusion, a genetic algorithm is introduced to automatically optimise the weights during feature fusion. At the initial stage, a number of weight sets are randomly generated, and the recognition accuracy obtained in facial recognition tasks by the fused features is used as the primary evaluation metric for the quality of weights. By iteratively performing operations such as selection, crossover, and mutation on the weights in the solution space, the quality of weights is optimized. After successive iterations, the optimal combination of weights is obtained, ensuring that the fused features exhibit optimal performance. Finally, employing the Euclidean distance as a measure of similarity between features, the fused features are classified using the nearest-neighbour classification method, and the recognition accuracy is computed. Experimental validation is conducted on the publicly available YALE and ORL datasets. The results indicate that, compared to other face recognition algorithms, this method exhibits superior robustness and higher recognition accuracy.

show abstract

Section: Feature Fusion Strategiesmentioning

confidence: 99%

Face Recognition Algorithm based on Optimal Fusion of LBP and HOG Features

2023

JISSR

View full text Add to dashboard Cite

show abstract

“…Furthermore, the fusion-based algorithms, both at the image level and the match score level, increased the recognition accuracy by at least 10.2-13.7% . Shahbe et al in [4], proposed a comparative study of several rank-based decision fusion techniques, specifically minimum ranking fusion, product ranking fusion, and average ranking fusion to fuse IR and visible modalities at match score level.The study, applied on the Equinox and OTCBVS databases, has shown a superiority of the average ranking technique over one modality based systems.…”

Section: Introductionmentioning

confidence: 99%

A comparative study of best spectral bands selection systems for face recognition

Bouchech

Foufou

2014

2014 IEEE/ACS 11th International Conference on Computer Systems and Applications (AICCSA)

View full text Add to dashboard Cite

Multispectral images (MI) have shown promising capabilities to solve problems resulting from high illumination variation in face recognition. However, the use of MI, with the huge number of captured spectral bands for each subject, is impractical unless a system for best spectral bands selection (BSBS) is used. In this work, first we give an up to date overview of the existing BSBS techniques proposed for face recognition. We aim to highlight the imporatnce of this component of MI based systems. The reviewed techniques are then experimented using the multispectral face database IRIS − M 3 to compare their performances. To the best of our knowledge this is the first study that reviews and compares existing techniques for BSBS. The Obtained results emphasized the importance of setting up techniques for BSBS with MI based systems as well as the need for new techniques that investigate better the increasing speed of image processing tools.

show abstract