Journal of Northeastern University Natural Science ›› 2017, Vol. 38 ›› Issue (3): 310-314.DOI: 10.12068/j.issn.1005-3026.2017.03.002

• Information & Control • Previous Articles     Next Articles

Face Recognition Based on Multi-feature Optimization Fusion of LBP, LPQ and Gabor with Multi-scale Blocks

JIA Ming-xing1, DU Jun-qiang1, SONG Peng-fei1, TIAN Shu2   

  1. 1. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China; 2. School of Software, Northeastern University, Shenyang 110169, China.
  • Received:2015-10-12 Revised:2015-10-12 Online:2017-03-15 Published:2017-03-24
  • Contact: JIA Ming-xing
  • About author:-
  • Supported by:
    -

Abstract: There is still a problem of low recognition rate in face recognition when dealing with complex situations such as illumination, pose and facial expression, so a face recognition method based on multi-feature optimization fusion with multi-scale blocks was proposed. Three complementary characters were considered as the features needed, such as local binary pattern(LBP), the local phase quantization(LPQ) and wavelet transform. Further, a method of multi-scale blocks of face grayscale image was proposed to consider that the single block algorithm made the surrounding information not be fully extracted and thereby lost useful features for face recognition. Finally, genetic algorithm was used to optimize the weights of multi-feature of multi-scale blocks, and the optimal weights could be obtained. Experiment 4 with this method was tested on the basis of large scale face data set FRGC2.0 database, and the validation rate of the method reached 95.31%, with recognition rate reaching 99.06%. Compared to the previous literature, this algorithm can extract more feature information of human face, and improve the face recognition rate.

Key words: LBP(local binary pattern), LPQ(local phase quantization), Gabor, optimal weight, multi-scale blocks, face recognition

CLC Number: