Palmprint recognition and palm vein recognition are two emerging biometrics technologies. In the past two decades, many traditional methods have been proposed for palmprint recognition and palm vein recognition, and have achieved impressive results. However, the research on deep learning-based palmprint recognition and palm vein recognition is still very preliminary. In this paper, in order to investigate the problem of deep learning based 2D and 3D palmprint recognition and palm vein recognition in-depth, we conduct performance evaluation of seventeen representative and classic convolutional neural networks (CNNs) on one 3D palmprint database, five 2D palmprint databases and two palm vein databases. A lot of experiments have been carried out in the conditions of different network structures, different learning rates, and different numbers of network layers. We have also conducted experiments on both separate data mode and mixed data mode. Experimental results show that these classic CNNs can achieve promising recognition results, and the recognition performance of recently proposed CNNs is better. Particularly, among classic CNNs, one of the recently proposed classic CNNs, i.e., EfficientNet achieves the best recognition accuracy. However, the recognition performance of classic CNNs is still slightly worse than that of some traditional recognition methods.
To simulate the firing pattern of biological grid cells, this paper presents an improved computational model of grid cells based on column structure. In this model, the displacement along different directions is processed by modulus operation, and the obtained remainder is associated with firing rate of grid cell. Compared with the original model, the improved parts include that: the base of modulus operation is changed, and the firing rate in firing field is encoded by Gaussian-like function. Simulation validates that the firing pattern generated by the improved computational model is more consistent with biological characteristic than original model. Besides, the firing pattern is badly influenced by the cumulative positioning error, but the computational model can also generate the regularly hexagonal firing pattern when the real-time positioning results are modified. 相似文献
针对蜂窝与终端直通(Device-to-device,D2D)混合网络中的不完全信道状态信息(Channel state information,CSI),提出了一种基于不完全CSI的最优功率分配算法。利用拉格朗日乘数法推导出了最优功率分配解的闭式表达式。相对于传统的最大发送功率分配方案,该最优解受信道信息误差的影响较小,具有较好的鲁棒性。仿真结果表明,本文算法可以节约发送功率,并得到更大的接收信噪比。 相似文献
Experimental tests show that static pre-loading has a significant effect on the dynamic strength of concrete.Based on meso-scale particle element model,numerical simulations of dynamic bending tests with pre-loading are performed.Complete stress–strain relationships are then obtained.Significant increase in dynamic strength is found when the pre-loadings are imposed within the elastic limit of concrete.However,when the imposition of pre-loadings reaches the plastic or softening range,dynamic strengths may gradually decrease along with the increase in pre-loadings.The distribution of energy components and the failure modes are discussed to explain the mechanisms of the phenomena. 相似文献
Based on the dual-phase-lagging(DPL)heat conduction model,the Cattaneo-Vernotte(CV)model and the improved CV model we investigate the one-dimensional heat conduction in gold films with nano-scale thickness exposed to an ultra-fast laser heating.The influence of system parameters on the temperature field is explored.We find that for all the non-Fourier heat conduction models considered in this work,a larger Knudsen number normally leads to a higher temperature.For the DPL model,the large ratio of the phase lag of temperature gradient to the phase lag of heat flux reduces the maximum temperature and shortens the time for the system to reach its steady state.The CV model and the improved CV model lead to the similar thermal wave behavior of the temperature field,but the thermal wave speeds for these two models are different,especially for large Knudsen numbers.When the phase lag of temperature gradient is smaller,the difference between the DPL model and the improved CV model is not significant,but for the large phase lag of temperature gradient the difference becomes quite significant,especially for the large Knudsen number.In addition,the effect of the surface accommodation coefficient,which is a parameter in the slip boundary condition,on the temperature field of the gold film heated by ultra-fast laser pulses is investigated based on the DPL model. 相似文献
Poly (vinyl butyral) (PVB) hollow fiber membranes were fabricated via thermally induced phase separation (TIPS). The effects of coagulation bath temperature (CBT) on the structure and performance of membranes were investigated in detail. The morphologies of the membranes were studied by scanning electron microscopy (SEM), the performances of water permeability, rejection, breaking strength and elongation were measured, respectively. The results indicate that all the membranes have the asymmetric morphology and the thickness of the skin layer decreases and the pore size of the outer layer increases with the increase of CBT. The permeability of membranes prepared at air gap 1.0 cm and take-up speed 0.253 m/s increases from 1.047×10-7 to 5.909×10-7 m3/(m2·s-kPa) with the CBT increasing from 20 ℃ to 40℃, and sharply increases to 35.226×10 7 m3/(m2.s.kPa)once the CBT arrives at 50 ℃. While the carbonic ink rejections have no significant decrease, totally exceed 98%, but that of acid-maleic acid copolymer greatly decreases with the increase of CBT. Both the breaking strength and elongation decrease with the increase of CBT. 相似文献