Search

With the growing demand for information security, finger-vein recognition has become widespread. However, the robustness of the recognition process becomes a major problem. When identifying unseen categories with traditional finger-vein recognition systems, a few issues remain, such as recognition interference and low efficiency. This paper proposes a Deep Generalized Label Finger-Vein (DGLFV) model to extract feature maps and achieve high-accuracy recognition. The largest rectangular finger-vein region is extracted through image semantic segmentation and the advanced bidirectional traversing and center diffusion method for the known categories. Then we generalize all the unseen categories actively as Class $C+1$ to reduce interference from unregistered users. Furthermore, an adaptive threshold acquisition algorithm is proposed for Label Receiver Operating Characteristic (LROC), so that the procedures of classification, recognition, and verification are unified. Apart from Shandong University Homologous Multi-modal Traits (SDUMLA-HMT), we have conducted additional experiments on our self-built database, Finger Veins of Signal and Information Processing Laboratory (FV-SIPL). The recognition accuracy of the approach proposed in this paper has reached 99.25% and 99.08% testing on FV-SIPL and SDUMLA-HMT, with a low error rate at 1.481% and 2.228% and little time consumption of 0.157s for a single image, which is better than most state-of-the-art finger-vein recognition methods.

The study of intelligent operation and maintenance methods for turbofan engines is of great importance for improving the reliability of turbofan engines. Given the harsh operating conditions and complex structure of the turbofan engine, it is extremely difficult to establish an accurate physical model for remaining useful life (RUL) prediction. The traditional operation and maintenance method based on the physical model has several limitations in the application of turbofan engines, while the data-driven method offers a new solution. Compared with traditional machine learning models, deep learning models possess more powerful nonlinear expression capabilities and feature extraction capabilities. Therefore, this study focuses on studying the RUL prediction algorithm for turbofan engines based on the fused deep learning models. In this article, a multimodal deep learning approach based on a 1DCNN (1D convolutional neural network) + attention enhanced Bi-LSTM (bidirectional long short-term memory) network is proposed to predict the RUL by mining the temporal information of data. Furthermore, a DDResNet (dilated deep residual network) is also introduced to the 1DCNN submodel to leverage its hidden pattern mining capability due to its advance in preventing performance degradation across layers. Subsequently, the output of these two submodels is weighted fused to obtain the final RUL prediction. The merits of the proposed method are demonstrated by comparing it with existing methods for RUL prediction using the C-MAPSS (commercial modular aero-propulsion system simulation) dataset.

The study aims to address the multiple pseudo-planes and low efficiency of random sample consensus (RANSAC) while extracting building planes from airborne light detection and ranging (LiDAR) data. A Naive Bayes–RANSAC method is proposed for the proportional segmentation of quantitative features. First, the point cloud is divided into ground and non-ground points using the cloth simulation filtering (CSF) algorithm. RANSAC is used to extract the non-ground point directly. Accordingly, several planes are obtained as the original building roofs. The differences in the density and echo features between the pseudo-plane and the training samples’ building plane are then analyzed. The percentage of the points whose densities are greater than the threshold, determined by the densities of the building edges, is calculated within each plane. Similarly, the percentage of the points whose echo numbers are higher than one is calculated within each plane. The two percentages calculated above are used as the classification basis of Naive Bayes. Finally, the building roofs can be extracted with the Naive Bayes classifier. This method can extract building roofs from LiDAR point clouds without external data. Two experiments are conducted to reveal the method’s practicality and reliability. The results demonstrate completeness of 89.0% and correctness of 94.0% at the per-area level for the first experiment and 90.7% and 91.7% for the second experiment. Simultaneously, the second experiment exhibits superior results compared with other methods under the same conditions.

Purpose The role of chemotherapy (CT) and radiotherapy (RT) for management of extraskeletal osteosarcoma (ESOS) remains controversial. We examined disease outcomes for ESOS patients and investigated the association between CT/RT with recurrence and survival. Patients and methods Retrospective review at 25 international sarcoma centers identified patients ≥18 years old treated for ESOS from 1971 to 2016. Patient/tumour characteristics, treatment, local/systemic recurrence, and survival data were collected. Kaplan–Meier survival and Cox proportional-hazards regression and cumulative incidence competing risks analysis were performed. Results 370 patients with localized ESOS treated definitively with surgery presented with mainly deep tumours (n = 294, 80%). 122 patients underwent surgical resection alone, 96 (26%) also received CT, 70 (19%) RT and 82 (22%) both adjuvants. Five-year survival for patients with localized ESOS was 56% (95% CI 51%–62%). Almost half of patients (n = 173, 47%) developed recurrence: local 9% (35/370), distant 28% (102/370) or both 10% (36/370). Considering death as a competing event, there was no significant difference in cumulative incidence of local or systemic recurrence between patients who received CT, RT, both or neither (local p = 0.50, systemic p = 0.69). Multiple regression Cox analysis showed a significant association between RT and decreased local recurrence (HR 0.46 [95% CI 0.26–0.80], p = 0.01). Conclusion Although the use of RT significantly decreased local recurrences, CT did not decrease the risk of systemic recurrence, and neither CT, nor RT nor both were associated with improved survival in patients with localized ESOS. Our results do not support the use of CT; however, adjuvant RT demonstrates benefit in patients with locally resectable ESOS.

Because of the large physical element size, there is significant difficulty in manufacturing of the planar ultrawideband modular antenna (PUMA) array operating in the P-L band. To solve this problem, a new PUMA element without capacitive coupling patches or capacitive through-vias is presented. The printed dipole in the PUMA element is asymmetric, and is fed through a vertical strip-line balun. Consequently, the designed PUMA element achieves ±45° scanning both in the E and H planes within a bandwidth of 6.5:1 in the P-L band. The sizes of the element are $50 \times 50\text{mm}(0.43 \lambda_{\mathrm{h}}, \lambda_{\mathrm{h}}$ is the free-space wavelength at the highest operating frequency), and the total profile is 76.5mm ( $0.66 \lambda_{\mathrm{h}}$ ). The simulation results of an infinite array show that the proposed element can scan up to ±45° in the E and H planes within 0.4–2.6GHz with orthogonal port isolation better than 10dB. The design is verified by measured results of an $8 \times 7$ array antenna prototype.

To address the potential airspace security threats of "low, slow, and small" targets, a dynamic Bayesian network model is used to determine the multi-target potential threat value’ and then establish the threat level. The target with the highest threat value is captured, and the improved A-Star algorithm is used. The size of the search space is reduced by combining with the maneuvering performance of the rotor-type UAV to improve the efficiency of real-time capture trajectory planning. Through simulation experiment analysis and verification, the multi-target threat value given by the dynamic Bayesian network was consistent with that in the expert experience evaluation. At the same time, the improved A-Star algorithm can give the optimal capture trajectory of the target with the greatest potential threat.

Small targets in long-distance aerial photography have the problems of small size and blurry appearance, and traditional object detection algorithms face great challenges in the field of small-object detection. With the collection of massive data in the information age, traditional object detection algorithms have been gradually replaced by deep learning algorithms and have an advantage. In this paper, the YOLOV3-Tiny backbone network is augmented by using the pyramid structure of image features to achieve multi-level feature fusion prediction. In order to eliminate the loss of spatial feature information and hierarchical information caused by pooling operations in convolution processes and multi-scale operations in multi-layer structures, a spatial attention mechanism based on residual structure is proposed. At the same time, the idea of reinforcement learning is introduced to guide bounding box regression on the basis of the rough positioning of the native boundary regression strategy, and the variable IoU calculation method is used as the evaluation index of the reward function, and the boundary regression model based on the reward mechanism is proposed for fine adjustment. The VisDrone2019 data set was selected as the experimental data support. Experimental results show that the mAP value of the improved small-object detection model is 33.15%, which is 11.07% higher than that of the native network model, and the boundary regression accuracy is improved by 23.74%.

The availability of paired examples greatly facilitates the task of style transfer by allowing the use of supervised learning. However, our scenario does not enjoy such a condition. We focus on style transfer for academic writing, and examine the possibility of performing style transfer between sentences from the abstract and conclusion sections of a scientific article in the Natural Language Processing field, in both directions. We assume a latent correlation between the abstract and conclusion styles, and construct an unpaired data set. We propose the use of a version of CydeGAN based on transformers to perform the task. Our approach is shown to realize differences in tense or word usage which are characteristic of the different sections.

This paper proposes a method to quantify the typicality of lexical bundles in sections of academic articles, specifically in the field of Natural Language Processing papers. Typicality is defined as the product of individual KL-divergence scores and the probability of a bundle to appear in a type of section. An evaluation of our typicality measure against two other baselines shows slight improvements according to the Silhouette coefficient.

Urban subway construction inevitably causes disturbances to the rock strata. It can even cause ground collapse accidents when construction encounters a bad geological body in the stratum. To verify the influence of cavities on surface settlement and the mechanism of formation instability, the instability mechanism of cavity-bearing strata under tunnel excavation disturbance was herein studied by tests using self-designed indoor models and numerical simulations. This study was based on the concrete project case of Qingdao Metro. Two groups of experiments (with and without cavities) were designed to simulate the four-step excavation by staged unloading of an airbag. The results show that the settlement value in the cavity state was about twice as much as that in the nonvoid state at the same stage. Besides the first step of excavation, in the numerical simulation, the settlement value of the same stage is about 1.3 times that of the model test. Simulating the deformation process of the surrounding rock in model tests shows that, in the excavation of the tunnel, the collapse surface of the tunnel arch roof will be connected with the sliding surface of the formation cavity. The cavity will gradually change from a regular circle to an ellipse or may even close, resulting in a sudden increase in land subsidence or even ground collapse. This indicates that disturbance during tunnel excavation would cause greater instability in a stratum when bad geological bodies, such as cavities, exist in the stratum.

Excessively high levels of nitrate in groundwater constitute an increasing threat to human health and environment, which is raising wide-spread concerns at the global scale. Although the monitoring targets merely on nitrate concentration in groundwater, the transfer of nitrate occurs through the whole internally connected vadose zone-groundwater system. Therefore, managers should address the groundwater nitrate problem from a surface-to-groundwater profile perspective, which, however, is relatively poorly understood. In this review, we systematically outline nitrate management measures including source reduction (source-targeted), in-situ groundwater remediation (receptor-targeted), and polluted groundwater reuse (resource-targeted). Furthermore, researchers have started incorporating the surface-to-groundwater profile view into developing the management decision support tools. Recent advances in the development of mathematical simulation models which coupled unsaturated and saturated zones and life cycle assessment-based multi-criteria decision analysis methods which considered the total environmental impacts are examined. These available measures and decision support tools have significant implications for selecting an optimal nitrate management strategy. Finally, we emphasize the necessity of creating a management framework from a surface-to-groundwater profile perspective and identify the directions of future research works and related challenges.

We reviewed the clinical characteristics and outcomes of patients treated for alveolar soft part sarcoma (ASPS) and analyzed the effect of surgery for patients presenting with and without metastatic disease (DM).The SEER Registry was queried for patients with ASPS from 1973-2012. The Kaplan-Meier estimate and Cox proportional hazards were used to analyze survival outcomes and risk variables.Among 251 patients, 43% had DM and 67% locoregional disease (LR) on presentation. The 5-year overall survival (OS) for all patients was 56% (82% and 27% for LR and DM, respectively). Multivariate analysis identified older age (hazard ratio [HR] = 1.03 per year, P 0.001), tumor size10 cm (HR = 2.76, P = 0.013), DM at diagnosis (HR = 3.79, P 0.001), and truncal primary site (HR = 1.63, P = 0.035) as independent factors predicting worse OS. For LR patients, surgery plus radiotherapy (RT) resulted in better OS compared to surgery alone P = 0.014. For DM patients, primary site surgery significantly improved survival (P 0.001).ASPS presents with high metastasis rate but has a relatively indolent clinical course and a favorable prognosis with prolonged survival. Aggressive treatment using adjuvant RT with surgery is indicated in patients with LR disease and surgery is indicated in patients presenting with DM. J. Surg. Oncol. 2016;113:581-586. © 2016 Wiley Periodicals, Inc.

A 70-year-old man with history of stage IV renal cell carcinoma, chronic atrial fibrillation on warfarin, coronary artery disease status post-percutaneous coronary intervention resulting in an ischaemic cardiomyopathy with left ventricular ejection fraction of 40%–45%, presented with shortness of breath 10 days after starting pazopanib. Within the first week of starting pazopanib, the patient developed fatigue and progressive dyspnoea on exertion. His symptoms quickly worsened and he had compromised mental status. He was transferred to the intensive care unit (ICU) and intubated due to continued respiratory distress. He was found to be in cardiogenic shock and was started on inotropic support with dobutamine and norepinephrine. With maximum support, the patient was slowly weaned off vasopressors and was successfully extubated on ICU day 9. His hospital stay lasted 29 days with management of multiple medical complications, and he was eventually discharged to a rehabilitation facility.

e23562Background: Retroperitoneal sarcomas (RPS) can cause local, intraperitoneal, and distant failures. Treatment requires a tailored, multidisciplinary strategy. Methods: In an IRB approved proto...