scholarly journals Statistical Evaluation of Emerging Feature Extraction Techniques for Aging-Invariant Face Recognition Systems

2017 ◽  
Vol 2 (1) ◽  
Author(s):  
Ayodele Oloyede ◽  
Temitayo Matthew Fagbola ◽  
Stephen Olabiyisi ◽  
Elijah Omidiora ◽  
John Oladosu
Keyword(s):  
Feature Extraction ◽  
Face Recognition ◽  
Wavelet Transform ◽  
Local Binary Pattern ◽  
Statistical Significance ◽  
Significance Test ◽  
Gabor Wavelet ◽  
Extraction Techniques ◽  
Statistical Significance Test ◽  
Recognition Systems

Large variation in facial appearances of the same individual makes most baseline Aging-Invariant Face Recognition Systems (AI-FRS) suffer from high automatic misclassification of faces. However, some Aging-Invariant Feature Extraction Techniques (AI-FET) for AI-FRS are emerging to help achieve good recognition results when compared to some baseline transformations in conditions with large amount of variations in facial texture and shape. However, the performance results of these AI-FET need to be further investigated statistically to avoid being misled. Statistical significance test can be used to logically justify such performance claims. The statistical significance test would serve as a decision rule to determine the degree of acceptability of the probability to make a wrong decision should such performance claims be found faulty. In this paper, the means between the quantitative results of emerging AI-FET (Histogram of Gradient (HoG), Principal Component Analysis-Linear Discriminant Analysis (PCA-LDA) and Local Binary Pattern-Gabor Wavelet Transform (LBP-GWT)) and the baseline aging-invariant techniques (Local Binary Pattern (LBP) and Gabor Wavelet Transform (GWT)) were computed and compared to determine if those means are statistically significantly different from each other using one-way Analysis of Variance (ANOVA). The ANOVA results obtained at 0.05 critical significance level indicate that the results of the emerging AI-FET techniques are not statistically significantly different from those of baseline techniques because the F-critical value was found to be greater than the value of the calculated F-statistics in all the evaluations conducted.

Pose and Illumination Invariant Hybrid Feature Extraction for Newborn

2019 ◽  
Vol 12 ◽  
Author(s):  
Rishav Singh ◽  
Ritika Singh ◽  
Aakriti Acharya ◽  
Shrikant Tiwari ◽  
Hari Om
Keyword(s):  
Feature Extraction ◽  
Face Recognition ◽  
Local Binary Pattern ◽  
Hybrid Approach ◽  
Controlled Environment ◽  
Illumination Invariant ◽  
Speeded Up Robust Features ◽  
Hybrid Feature Extraction ◽  
Recognition Systems ◽  
Gallery Image

Recently a lot of face recognition systems are being designed to identify individuals in a semi controlled environment where pose and illumination are controlled. However, in the case of newborns it is not easy to click the photographs with similar pose and illumination. Here, in this paper a hybrid approach using Speeded Up Robust Features (SURF) and Local Binary Pattern (LBP) is proposed for newborns. Moreover, in this paper the experiment is done for a single gallery image with improved results. It shows that the proposed method has 97.18% accuracy which is an 8% improvement over LBP and 8.6% improvement over SURF for Rank 5.

A Feature Extraction Method based on Local Binary Pattern Preprocessing and Wavelet Transform

2020 ◽  
Vol 34 (13) ◽  
pp. 2050030
Author(s):  
Peng-Yi Liu ◽  
Zhi-Ming Li
Keyword(s):  
Pattern Recognition ◽  
Feature Extraction ◽  
Face Recognition ◽  
Wavelet Transform ◽  
Extraction Method ◽  
Local Binary Pattern ◽  
Face Image ◽  
Local Descriptors ◽  
Feature Extraction Method

Face recognition has been extensively studied by many scholars in the recent decades. Local binary pattern (LBP) is one of the most popular local descriptors and has been widely applied to face recognition. Wavelet transform is also more and more active in the field of pattern recognition. In this paper, a novel feature extraction method is proposed to overcome illumination influence. First, a given face image is processed by the LBP operator, and an LBP image is obtained. Second, wavelet transform is used to extract discriminant feature from the LBP image. The experiment results on LFW, Extended YaleB and CMU-PIE face databases show that the proposed method outperforms several popular face recognition methods, and the preprocessing step plays an important role to extract effective features for classification.

Gabor-scale binary pattern for face recognition

2016 ◽  
Vol 14 (05) ◽  
pp. 1650035 ◽  
Author(s):  
Zhi-Ming Li ◽  
Zheng-Hai Huang ◽  
Ting Zhang
Keyword(s):  
Face Recognition ◽  
Wavelet Transform ◽  
Spatial Frequency ◽  
Local Binary Pattern ◽  
Experimental Results ◽  
Gabor Wavelet ◽  
Special Orientation ◽  
Scale Ratio ◽  
Face Descriptor ◽  
Scale Length

In this paper, a novel face descriptor, the Gabor-scale binary pattern (GSBP), is proposed to explore the neighboring relationship in spatial, frequency and orientation domains for the purpose of face recognition. In order to extract the GSBP feature, the Gabor-scale volume and the Gabor-scale vector are introduced by using a group of Gabor wavelet coefficients with a special orientation. Moreover, the Gabor-scale length pattern and the Gabor-scale ratio pattern are proposed. Compared with the existed methods, GSBP utilizes the deep relations between neighboring Gabor subimages instead of directly combining Gabor wavelet transform and local binary pattern. For estimating the performance of GSBP, we compare the proposed method with the related methods on several popular face databases, including LFW, FERET, AR, Yale and Extended YaleB databases. The experimental results show that the proposed method outperforms several popular face recognition methods.

scholarly journals PEM-PCA: A Parallel Expectation-Maximization PCA Face Recognition Architecture

2014 ◽  
Vol 2014 ◽  
pp. 1-16 ◽  
Author(s):  
Kanokmon Rujirakul ◽  
Chakchai So-In ◽  
Banchar Arnonkijpanich
Keyword(s):  
Feature Extraction ◽  
Face Recognition ◽  
Expectation Maximization ◽  
High Speed ◽  
Expectation Maximization Algorithm ◽  
Parallel Architecture ◽  
Principal Component ◽  
Eigenvalue Decomposition ◽  
Computational Time ◽  
Recognition Systems

Principal component analysis or PCA has been traditionally used as one of the feature extraction techniques in face recognition systems yielding high accuracy when requiring a small number of features. However, the covariance matrix and eigenvalue decomposition stages cause high computational complexity, especially for a large database. Thus, this research presents an alternative approach utilizing an Expectation-Maximization algorithm to reduce the determinant matrix manipulation resulting in the reduction of the stages’ complexity. To improve the computational time, a novel parallel architecture was employed to utilize the benefits of parallelization of matrix computation during feature extraction and classification stages including parallel preprocessing, and their combinations, so-called a Parallel Expectation-Maximization PCA architecture. Comparing to a traditional PCA and its derivatives, the results indicate lower complexity with an insignificant difference in recognition precision leading to high speed face recognition systems, that is, the speed-up over nine and three times over PCA and Parallel PCA.

Sign in / Sign up

Export Citation Format

Share Document