基于数据质量评价的目标估计

针对目标估计过程需要大量人工参与、自动化程度低的问题，提出了基于数据质量评价的目标估计方法。利用目标数据质量评价方法，对不同传感器得到的目标数据质量进行科学、有效的测度和评价，并根据质量得分动态调整各数据源在目标估计过程中所占的权重，从而减少人工干预，提高目标估计效能。仿真试验结果证明了该方法的有效性。     

基于单光子计数激光雷达的Bayesian检测研究

在单光子计数激光雷达检测领域,目前的检测方法在低信噪比情况下虚警概率会增加,同时也无法适应噪声变化的问题。针对这些问题,提出了一种基于Bayesian的检测方法,该方法首先通过雷达方程估计回波信号光子数的范围,将其作为先验信息,而后结合二项分布建立了累计概率模型,基于Bayesian判决准则计算得到检测阈值,此阈值能够在检测概率与虚警概率中间择其平衡。这种方法不仅克服了低信噪比检测困难的情况,还减少了先验信息的获取难度。实验结果表明,对比固定阈值其虚警概率降低了10倍。对比“恒虚警”其检测概率提高了约20。验证了方法具有良好的检测效果,具备一定的可操作性。     

机器翻译译文质量估计综述

机器翻译译文质量估计(Quality Estimation,QE)是指在不需要人工参考译文的条件下,估计机器翻译系统产生的译文的质量,对机器翻译研究和应用具有很重要的价值。机器翻译译文质量估计经过最近几年的发展,取得了丰富的研究成果。该文首先介绍了机器翻译译文质量估计的背景与意义;然后详细介绍了句子级QE、单词级QE、文档级QE的具体任务目标、评价指标等内容,进一步概括了QE方法发展的三个阶段: 基于特征工程和机器学习的QE方法阶段,基于深度学习的QE方法阶段,融入预训练模型的QE方法阶段,并介绍了每一阶段中的代表性研究工作;最后分析了目前的研究现状及不足,并对未来QE方法的研究及发展方向进行了展望。     

Presentation attack detection based on two-stream vision transformers with self-attention fusion

Aiming at the performance degradation of the existing presentation attack detection methods due to the illumination variation, a two-stream vision transformers framework (TSViT) based on transfer learning in two complementary spaces is proposed in this paper. The face images of RGB color space and multi-scale retinex with color restoration (MSRCR) space are fed to TSViT to learn the distinguishing features of presentation attack detection. To effectively fuse features from two sources (RGB color space images and MSRCR images), a feature fusion method based on self-attention is built, which can effectively capture the complementarity of two features. Experiments and analysis on Oulu-NPU, CASIA-MFSD, and Replay-Attack databases show that it outperforms most existing methods in intra-database testing and achieves good generalization performance in cross-database testing.     

Decomposition and replacement: Spatial knowledge distillation for monocular depth estimation

Knowledge distillation has become a key technique for making smart and light-weight networks through model compression and transfer learning. Unlike previous methods that applied knowledge distillation to the classification task, we propose to exploit the decomposition-and-replacement based distillation scheme for depth estimation from a single RGB color image. To do this, Laplacian pyramid-based knowledge distillation is firstly presented in this paper. The key idea of the proposed method is to transfer the rich knowledge of the scene depth, which is well encoded through the teacher network, to the student network in a structured way by decomposing it into the global context and local details. This is fairly desirable for the student network to restore the depth layout more accurately with limited resources. Moreover, we also propose a new guidance concept for knowledge distillation, so-called ReplaceBlock, which replaces blocks randomly selected in the decoded feature of the student network with those of the teacher network. Our ReplaceBlock gives a smoothing effect in learning the feature distribution of the teacher network by considering the spatial contiguity in the feature space. This process is also helpful to clearly restore the depth layout without the significant computational cost. Based on various experimental results on benchmark datasets, the effectiveness of our distillation scheme for monocular depth estimation is demonstrated in details. The code and model are publicly available at : https://github.com/tjqansthd/Lap_Rep_KD_Depth.     

Predictive process mapping for laser powder bed fusion: A review of existing analytical solutions

One of the main challenges in the laser powder bed fusion (LPBF) process is making dense and defect-free components. These porosity defects are dependent upon the melt pool geometry and the processing conditions. Power-velocity (PV) processing maps can aid in visualizing the effects of LPBF processing variables and mapping different defect regimes such as lack-of-fusion, under-melting, balling, and keyholing. This work presents an assessment of existing analytical equations and models that provide an estimate of the melt pool geometry as a function of material properties. The melt pool equations are then combined with defect criteria to provide a quick approximation of the PV processing maps for a variety of materials. Finally, the predictions of these processing maps are compared with experimental data from the literature. The predictive processing maps can be computed quickly and can be coupled with dimensionless numbers and high-throughput (HT) experiments for validation. The present work provides a boundary framework for designing the optimal processing parameters for new metals and alloys based on existing analytical solutions.     

A comparison of shear stress estimation methods for a single geobag on a rough bed

There are several methods for estimating bed shear stress in the literature, but comprehensive comparisons among them are limited and under specific conditions. This study compared these methods first on a bare smooth bed, and then for a single geobag on a rough bed in the interest of determining the stability of geobags used in riverbank protection structures. The geobag was filled with cement or sand and tested under different open channel flow conditions. The turbulent kinetic energy method appeared to best represent the local bed shear stress on the geobag when using the newly calibrated proportionality constants. The Reynolds stress method via extrapolation was relatively unaffected by changes to the geobags shape and measurement locations, suggesting this method inadequately represents the local bed shear stress. The Patel method and the universal law of the wall method failed to represent local bed shear stress in the rough bed cases due to instrument limitations and the breakdown of the law of the wall. This study highlights the impact of different methods on the bed shear stress estimation.     

Convolutional Neural Network Auto Encoder Channel Estimation Algorithm in MIMO-OFDM System

Higher transmission rate is one of the technological features of prominently used wireless communication namely Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing (MIMO–OFDM). One among an effective solution for channel estimation in wireless communication system, specifically in different environments is Deep Learning (DL) method. This research greatly utilizes channel estimator on the basis of Convolutional Neural Network Auto Encoder (CNNAE) classifier for MIMO-OFDM systems. A CNNAE classifier is one among Deep Learning (DL) algorithm, in which video signal is fed as input by allotting significant learnable weights and biases in various aspects/objects for video signal and capable of differentiating from one another. Improved performances are achieved by using CNNAE based channel estimation, in which extension is done for channel selection as well as achieve enhanced performances numerically, when compared with conventional estimators in quite a lot of scenarios. Considering reduction in number of parameters involved and re-usability of weights, CNNAE based channel estimation is quite suitable and properly fits to the video signal. CNNAE classifier weights updation are done with minimized Signal to Noise Ratio (SNR), Bit Error Rate (BER) and Mean Square Error (MSE).