The accuracy of retinal vessels segmentation is of great significance for the diagnosis of cardiovascular diseases such as diabetes and hypertension. Especially, the segmentation accuracy of the end of vessels will be affected by the area outside the retinal in fundus image. In this paper, we propose an attention guided U-Net with atrous convolution(AA-UNet), which guides the model to separate vessel and non-vessel pixels and reuses deep features. Firstly, AA-UNet regresses a boundary box to the retinal region to generate an attention mask, which was used as a weighting function to multiply the differential feature map in the model to make the model pay more attention to the vessels region. Secondly, atrous convolution replaces ordinary convolution in feature layer, which can increase the receptive field and reduce the amount of computation. Then, we add two shortcuts to the atrous convolution in order to reuse the features, so that the details of vessel are more prominent. We test our model with the accuracy are 0.9558/0.9640/0.9608 and AUC are 0.9847/0.9824/0.9865 on DRIVE, STARE and CHASE_DB1 datasets, respectively. The results show that our method has improvement in the accuracy of retinal vessels segmentation, and exceeded other representative retinal vessels segmentation methods.

Multimodal biometric system has received increasing interest as it offers a more secure and accurate authentication solution than unimodal systems. However, existing biometric fusion methods are still inadequate in dealing with correlations and redundancy of multimodal features simultaneously, causing bottlenecks in performance improvement. To overcome the above problem, this paper proposes an end-to-end multimodal finger recognition model that incorporates attention mechanisms into a similarity-aware encoder to achieve accurate recognition results for accurate identification results. Firstly, due to the different distribution of fingerprint and finger vein images, we propose a finger asymmetric backbone network (FAB-Net) for extracting discriminative intra-modal features, which reduces the network width by efficient utilization of feature maps. Then, a novel attention-based encoder fusion network (AEF-Net) with fused similarity performs dimensionality reduction-based fusion on multimodal multilevel features to alleviate performance degradation due to information redundancy. We also introduce channel attention in AEF-Net, which differs from the traditional attention mechanism by considering interdependencies between modalities to further improve performance. Extensive recognition experiments are conducted on three multimodal finger databases to verify the effectiveness of our method compared to state-of-the-art methods. Detailed ablation studies have also been carried out, which demonstrated that encoder-based reconstruction of redundant information can improve recognition performance.

Compared to other biometric features, finger vein depict vascular topology including some long-distance vascular and local textures, and have received much attention for their advantages in liveness detection and convenience of collection. However, some recent studies have focused on global-based spatial attention mechanisms to enhance the capture of long-distance information, but ignore local texture information, which is usually computationally inefficient. To address this problem, we propose an Axially-enhanced Local Attention Network (ALA Net) that uses the distribution of vascular topology to infer attention. Specifically, the proposed Axially-enhanced Local Attention (ALA) locates the receptive domain of each representation to its deformable neighborhood, thus extracting the vein pattern more flexibly and accurately. In addition, we propose a feature amplification strategy using a low-cost group convolution to improve the feature similarity and characterization accuracy between channels, resulting in higher accuracy at lower parameter scales. The effectiveness of the method is illustrated by searching the acceptance domain of the ALA module using image distribution with the accuracy of 94.50% on the SDUMLA dataset, 97.86% on the HKPU dataset, and 98.92% on the NUPT dataset, and the advancement of the ALA Net is demonstrated by comparing it with the state-of-the-art attention model and backbone network.