Search results
Source code for the 2019 IEEE TIFS paper "PalmNet: Gabor-PCA Convolutional Networks for Touchless Palmprint Recognition" - AngeloUNIMI/PalmNet
Apr 14, 2019 · Touchless palmprint recognition systems enable high-accuracy recognition of individuals through less-constrained and highly usable procedures that do not require the contact of the palm with a surface. To perform this recognition, methods based on local texture descriptors and convolutional neural networks (CNNs) are currently used to extract highly discriminative features while compensating ...
- A. Genovese, V. Piuri, K. N. Plataniotis, F. Scotti
- 2019
The Philadelphia Area Labor-Management Committee (PALM) is a private non-profit organization founded by the Philadelphia Council of the AFL-CIO and the Greater Philadelphia Chamber of Commerce in 1982. The mission of the organization is to foster increased labor management cooperation in the Southern Pennsylvania, Bucks, Chester, Delaware ...
- Gabor amplitude-phase Model
- Backbone Network
- Head Network
In the classical Gabor Amplitude-Phase model, the 2D complex Gabor filter is introduced. Its form is where, \( u =x \cos \theta +y \sin \theta , v =-x \sin \theta +y \cos \theta \), and \(\omega , \theta \) denote the frequency and orientation of the filter, respectively, which could be pre-calculated in Sect. 3.2. \(\sigma _{\theta }, \sigma _{\th...
As the network deepens, we define texture feature (TF), orientation feature (OF), and quality feature (QF) according to the semantics. The primary layer of the backbone network is stacked by residual blocks , and we implement convolution with a stride to replace pooling operation. Orientation Field Module. Standard ridge orientation range is [0\(^\...
After the backbone network converges, we freeze backbone weights and start training the head network. As we state in Sect. 3.1, amplitude image contains image quality information and the phase image reflects image texture information. In our next experiment, we use the complementary characteristics and extract minutiae based on these two images, an...
PalmNet Matlab source code for the paper: A. Genovese, V. Piuri, K. N. Plataniotis, and F. Scotti, "PalmNet: Gabor-PCA Convolutional Networks for Touchless Palmprint Recognition", IEEE Transactions on Information Forensics and Security, December 2019.
Source code for the 2019 IEEE TIFS paper "PalmNet: Gabor-PCA Convolutional Networks for Touchless Palmprint Recognition" - GitHub - xingbod/PalmNet_Angelo: Source code for the 2019 IEEE TIFS paper "PalmNet: Gabor-PCA Convolutional Networks for Touchless Palmprint Recognition"
PalmNet is a novel method of applying Gabor filters in a CNN to extract palmprint-specific descriptors and adapt to heterogeneous databases. It also fuses palmprint and inner finger texture features for higher accuracy without requiring class labels or pretrained filters.
