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。     

一种基于FCOS神经网络的小建筑物目标检测方法北大核心

提出了一种基于FCOS神经网络的小建筑物目标检测算法,针对FCOS模型在特征提取阶段提取到的小建筑物目标特征较少问题,引入多尺度检测和可变形卷积方式,加强网络对小建筑物目标的特征提取能力,并通过改进后的SGE注意力机制降低特征图中的干扰噪声权重。改进后的网络可以提取到更多的小建筑物目标特征,对环境干扰噪声的鲁棒性更强。在自己搭建的数据集上进行了实验测试,结果表明,在相同环境下网络改进后建筑物的整体检测准确率提升了1.7%,其中对小建筑物目标提升了3.6%,减少了小建筑物目标漏检、误检的问题。     

成型温度对PBX装药内部质量及性能的影响

针对压制成型的PBX炸药装药，选择CT无损检测、巴西实验和扫描电镜检测等技术，对比研究了室温和加热两种温度下压制成型的炸药装药内部质量、静态力学性能和细观破坏形式。结果表明，加热压制有利于改善炸药装药的内部质量，可避免产生初始损伤，且提高了装药的力学性能。细观尺度上，室温压制成型的装药主要发生界面脱黏破坏，加热压制成型装药的主要破坏形式是穿晶断裂。     

一种直升机避障电力线检测方法

电力线是一类形状细长、特征稀疏、随着视角的变化容易混淆在大量背景信息中的特殊障碍物，常规电力线检测识别算法得到的目标框对电力线所在位置的估计不够准确。为此，提出了一种相对角度估计方法，基于常规电力线目标检测与识别算法，并结合电力线相对角度估计，从而提高电力线的检测识别过程中所在位置的精度。相比电力线绝对角度回归的方法，提出的相对角度估计方法容易训练易收敛，计算量小，适用于实时性要求较高的应用场合。     

基于误差估值累加开环校正的诱导式欺骗检测方法

诱导式卫星欺骗干扰可诱导航空器逐渐偏离预定航迹，难以被发现，因此及时有效地检测干扰是飞行安全的保障。在现有紧组合导航体制基础上，设计了一种基于误差估值累加开环校正的紧组合导航结构，并证明了其性能与传统闭环校正紧组合导航性能等效。在此结构中，将紧组合导航系统与自适应序贯概率比检测方法结合，提出了一种基于误差估值累加开环校正的诱导式欺骗检测方法，融合紧组合导航信息与其他不受欺骗影响的导航信息，构建欺骗检测统计量进行诱导式欺骗检测。仿真结果表明，开环校正结构可避免随时间累加的惯性导航系统误差所导致的组合导航滤波器发散问题，同时欺骗检测方法可进一步提高算法对“最坏”情形下微小诱导式欺骗的检测效果。     

基于SSD的眨眼与打哈欠检测

提出了一种新的基于SSD网络的眨眼与打哈欠检测。首先,对原SSD架构进行四个方面的修改。第一,主干网用MobileNetV3替换了VGG,使得架构尺寸缩减三分之一。第二,重新设计了输出检测器,实现了目标检测与分类、目标定位与高宽比回归多任务的融合。第三,网络直接回归高宽比,比传统的从关键点离线计算高宽比的方法鲁棒性要高。第四,省去了非最大抑制层(NMS)中的重叠度(IOU)计算和阈值设定,使得单目标检测速度提高了40%。另外,在数据集YawDD上验证了所提出方法的有效性,眨眼与打哈欠的测全率为95.1%与100.0%,在单卡2080Ti上实现了41fps的高检测速度。     

基于Massive MIMO及频谱叠加的5G SA网络上行优化方法

大规模多输入多输出(Multiple-Input Multiple-Output, MIMO)可以有效提升5G SA网络的上行链路数据传输速率以及可靠性。针对5G SA网络上行链路速率和覆盖不均衡的情况,提出了基于大规模MIMO的分组算法,将发送信号矢量进行分组,组内采用最大似然检测,组外采用基于正交三角分解(QR分解)的干扰消除检测,并且结合5G频谱的叠加策略,在降低算法复杂度的同时,有效提升网络覆盖和速率。通过5G SA现网实测,通过MIMO降低分组数量能够提升分组检测性能,结合上行低频段频谱叠加策略能够有效提升5G SA网络上行覆盖30%,提升5G SA网络上行平均速率40%~80%,特别是弱覆盖边缘的网络速率,最高可达600%。     

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.