A simple and fast representation-based face recognition method

A nearest neighbor classifier based on virtual test samples for face recognition

“…Then the training samples would have 21( ) different combinations. And the rate of classification errors of our method, an improvement to the nearest neighbor classifier method in [9], SRMVS, a simple and fast sparse representation method in [12] and TPTSR are , , , , .The Table 3 shows the experiment results which achieved a better rate of classification of the proposed method. It implies that the proposed method is more advantageous in small samples FR applications.…”

“…The training samples have been taken from 1 to 3 in the experiment. Table 1 And when the number of training samples was equal to 2, the rate of classification errors of our method, an improvement to the nearest neighbor classifier method in [9], SRMVS, a simple and fast sparse representation method in [12] and TPTSR are , , , , . It can be inferred from the results that the proposed method is better than other methods and the virtual test samples is helpful on the small A c c e p t e d M a n u s c r i p t samples.…”

“…Then the training samples would have 11( ) different combinations. When only one image is selected as the training sample, the rate of classification errors of our method, an improvement to the nearest neighbor classifier method in [9], SRMVS, a simple and fast sparse representation method in [12] and TPTSR are , , ,…”

Human Identification via Face Recognition: Comparative Study

“…Now adays human identification via face recognition methods are well known issues and have a wide range of applications [1], [2].Face recognition is an attractive filed that combining of image processing, statistical measures, pattern recognition and computer vision and other related fields [3], [4], [5].Different methods of face recognition are designed and implemented such as discrete wavelet transform [6], Curvelet Transform [7], discrete cosine transform [8], principal component analysis [9], independent component analysis [10].Face recognition is an important field of Image processing, and it play an important part of security field [11], [12], [13].Face image representation using different techniques are applied [14], [15].Image Fusion methods for classification and extraction are used to reconstruct the image [16], [17], [18].Fusion technique can be applied for images and videos [19], [20].Quality improving of image is a good issue and it is required to reduce the error as possible and reaching high accuracy [21], [22], [23].Face recognition and enhancement are implemented via different methods [24], [25].Image watermarking leads to security aspect of face identification [26], [27], [28].After this brief introduction we can say that there are many techniques (DWT, DFT, DCT, PCA, SVM …etc.) used for face recognition and the accuracy of these technique depends of many factors such as type of data set, focusing, lighting, resolution, … etc.…”

Robust coarse-to-fine face recognition method

“…LRC defines the task of face recognition as a problem of linear regression. After the LRC, Xu et al propose coarse-to-fine face recognition (CFFR) classifier [27] and alternative scheme of coarse-to-fine face recognition (ASCFFR) method, which both choose some classes in the first phase and uses the chosen classes to classify the test sample.…”