Face Recognition with Multi-Resolution Spectral Feature Images

Sun, Zhan-Li; Lam, Kin-Man; Dong, Zhao-Yang; Wang, Han; Gao, Qing-Wei; Zheng, Chun-Hou
February 2013
PLoS ONE;Feb2013, Vol. 8 Issue 2, p1
Academic Journal
The one-sample-per-person problem has become an active research topic for face recognition in recent years because of its challenges and significance for real-world applications. However, achieving relatively higher recognition accuracy is still a difficult problem due to, usually, too few training samples being available and variations of illumination and expression. To alleviate the negative effects caused by these unfavorable factors, in this paper we propose a more accurate spectral feature image-based 2DLDA (two-dimensional linear discriminant analysis) ensemble algorithm for face recognition, with one sample image per person. In our algorithm, multi-resolution spectral feature images are constructed to represent the face images; this can greatly enlarge the training set. The proposed method is inspired by our finding that, among these spectral feature images, features extracted from some orientations and scales using 2DLDA are not sensitive to variations of illumination and expression. In order to maintain the positive characteristics of these filters and to make correct category assignments, the strategy of classifier committee learning (CCL) is designed to combine the results obtained from different spectral feature images. Using the above strategies, the negative effects caused by those unfavorable factors can be alleviated efficiently in face recognition. Experimental results on the standard databases demonstrate the feasibility and efficiency of the proposed method.


Related Articles

  • Face Recognition via Local Directional Pattern. Dong-Ju Kim; Sang-Heon Lee; Myoung-Kyu Sohn // International Journal of Security & Its Applications;Mar2013, Vol. 7 Issue 2, p191 

    In this paper, we propose an illumination-robust face recognition system via local directional pattern images. Usually, local pattern descriptors including local binary pattern and local directional pattern have been used in the field of the face recognition and facial expression recognition,...

  • Embedded Face Detection and Recognition. Günlü, Göksel // International Journal of Advanced Robotic Systems;Oct2012, Vol. 9, p1 

    The need to increase security in open or public spaces has in turn given rise to the requirement to monitor these spaces and analyse those images on-site and on-time. At this point, the use of smart cameras - of which the popularity has been increasing - is one step ahead. With sensors and...

  • A Novel Algorithm for DOA Estimation of Multi-path Channel. Fugang Liu; Ming Diao // International Proceedings of Computer Science & Information Tech;2012, Vol. 50, p164 

    By analyzing the spatial smoothing MUSIC algorithm, we propose the transient smoothing algorithm and apply it in the presence of multi-path channel. It can be used in the estimation of DOA in Eigen-space. The relation between ratio of phase and time and the position of transmitter has been...

  • Fast Face Detection Based on AdaBoost and Canny Operators. Lin Zhang // Applied Mechanics & Materials;2014, Vol. 716-717, p936 

    Detection speed of traditional face detection method based on AdaBoost algorithm is slow since AdaBoost asks a large number of features. Therefore, to address this shortcoming, we proposed a fast face detection method based on AdaBoost and canny operators in this paper. Firstly, we use canny...

  • High Density Impulse Noise Removal Using Robust Estimation Based Filter. Vijaykumar, V. R.; Vanathi, P. T.; Kanagasabapathy, P.; Ebenezer, D. // IAENG International Journal of Computer Science;Sep2008, Vol. 35 Issue 3, p259 

    In this paper a novel method for removing fixed value impulse noise using robust estimation based filter is proposed. The function of the proposed filter is to detect the outlier pixels and restore the original value using robust estimation. Comparison shows the proposed filter effectively...

  • Acceleration of Genetic Algorithm for Peak Power Reduction of OFDM Signal. Shigei, Noritaka; Miyajima, Hiromi; Ozono, Keisuke; Araki, Kentaro // IAENG International Journal of Computer Science;Mar2011, Vol. 38 Issue 1, p32 

    Orthogonal frequency division multiplexing (OFDM) is superior in spectral efficiency and is widely used in today's digital communication. One of the drawbacks of OFDM is that the peak-to-average power ratio (PAPR) of the transmitted signal tends to be high. In order to overcome this problem,...

  • Signal-Based Performance Evaluation of Dereverberation Algorithms. Naylor, Patrick A.; Gaubitch, Nikolay D.; Habets, Emanu¨el A. P. // Journal of Electrical & Computer Engineering;2010, p1 

    We address the measurement of reverberation in terms of the (DRR) in the context of the assessment of dereverberation algorithms for which we wish to quantify the level of reverberation before and after processing. The DRR is normally calculated from the impulse response of the reverberating...

  • Blind Detection Algorithm for Frequency Hopping Signals Using Texture Feature. LV Chen-jie; WANG Bin // Journal of Signal Processing;Apr2015, Vol. 31 Issue 4, p453 

    In order to detect the frequency hopping signals at the communication countermeasures, a texture feature based algorithm is proposed. In this algorithm the received signals are represented as time-frequency diagram, and texture features are extracted from the time-frequency diagram by using gray...

  • Improve the Quality of the Edge and the Motion Estimation of Super-resolution Image Reconstruction. LIANG Yu; WANG Rui; LI Rui // Journal of Signal Processing;Jan2015, Vol. 31 Issue 1, p26 

    The projection onto convex sets algorithm is widely used for super-resolution image reconstruction. To address the fuzzy edge problems arising from conventional POCS super-resolution image reconstruction algorithm, four edge directions of 0°, 45°, 90°, 135° were detected first...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics