Support Vector Machines for 3D Object Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Orthonormal ridgelets and linear singularities
SIAM Journal on Mathematical Analysis
Online Palmprint Identification
IEEE Transactions on Pattern Analysis and Machine Intelligence
Digital curvelet transform for palmprint recognition
SINOBIOMETRICS'04 Proceedings of the 5th Chinese conference on Advances in Biometric Person Authentication
A novel method for palmprint feature extraction based on modified pulse-coupled neural network
ICIC'13 Proceedings of the 9th international conference on Intelligent Computing Theories and Technology
Hi-index | 0.00 |
A multi-scale palm print classification method based on FRIT (finite ridgelet transform) and SVM (support vector machine) was proposed. First, palm image with preprocessing was decomposed by using FRIT. As a result, ridgelet coefficients in different scales and various angels were obtained. The important linear feature of palmprint was included in the low frequency coefficients of FRIT decomposition coefficients. After the decomposition coefficients were transformed into feature vectors, SVM was chosen as a classifier. These feature vectors were regarded as feature parameters of palm print and sent into SVM to training. Two kernel functions are used as a discriminant function. Finally, SVM trained was used for classification of palmprint. The experiments were performed in PolyU Palmprint Database. The results indicate that proposed method has better performance than wavelet with SVM method, and classification accuracy used RBF (radial basis function) as kernel is higher than the use of polynomial kernel function.